Including Risk Part 1 Non Adaptive. Bruce A. McCarl


 Tracy Montgomery
 1 years ago
 Views:
Transcription
1 Including Ris Part Non Adaptive Bruce A. McCarl Specialist in Applied Optimization Professor of Agricultural Economics, Texas A&M Principal, McCarl and Associates agecon.tamu.edu/faculty/mccarl B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
2 Including Non Adaptive Ris Many face pervasive uncertainty. Prices, weather, labor, and other factors induce large changes in revenue, yields, woring rates and resources available. When incorporating ris into the program models there are three big issues. What is the nature of ris? a. What parameters of the model are uncertain? and b. How do we describe their distribution?. When during the model time horizon are ris outcomes revealed? Are there times when the model should reflect that the producer has received information about uncertain events and will mae adaptive decisions?. How do we model the decision maers behavioral reaction to ris? Is expected profit maximization not to the proper obective but rather some degree of aversion to the variation caused by ris? B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
3 Including Non Adaptive Ris Our ris treatment will be limited and somewhat specialized because of time constraints (see newboo.pdf chapter and probab.pdf for more extensive treatment) The treatment will be limited in several principal ways. Will specialize in ris in obective function coefficients. Will not discuss how to form such probability distributions except through a few casual remars (see probab.pdf for more extensive treatment). We will only cover the expected value variance formulation of model obective function alterations to ris (this is the two main one used in the literature  see newboo.pdf chapter for more extensive treatment). We will treat non adaptive behavior here. In a later section adaptive behavior using discrete stochastic programming or stochastic programming with recourse B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
4 Including Non Adaptive Ris Why model ris Why not ust solve for all values of risy parameters Curses of dimensionality and certainty Dimensionality Number of possible plans ( possible values for 5 parameters 5 = ) Certainty Each plan would be certain of data so we would have different things we could do What would we do? General Ris Modeling Aim Generate a plan which is Robust in the face of the Uncertainty Not best performer necessarily in any setting, but a good performer across many or most of the uncertainty spectrum B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
5 Including Non Adaptive Ris Ris entry into a programming model Maximize C Subect to A # b $ 0 Obective function returns  C Variability in prices Variability in production quantities Variability in costs Variability in maret sales Resource usages  A Variability in raw input quality Variability in woring conditions Variability in intermediate product yields Variability in product requirements Resource endowments  b Variability in demand firm faces Variability in resources available Variability in woring conditions B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 5
6 Including Non Adaptive Ris Forms of assumed reaction to ris NonRecourse or non adaptive decision maing Decisions made now consequences felt later No additional decisions made between now and when consequences felt Example Buy stoc now mae no decisions for one year Recourse or adaptive decisionmaing Decisions made now consequences arise over time Later time during model additional decisions made. In this later decision period Decision maer nows what happened between first decision and now. Decision maers cannot revise prior actions but can adust current decisions ie current decisions can be employed to mae adustments in the face of realized events  phenomena called irreversibility and recourse Example Buy stoc now, review decisions quarterly possibly selling and buying other stocs B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 6
7 Including Non Adaptive Ris Decision Maer reaction to ris Expected Value Maximization max c s. t. A b o Conservative  Fat or thin coefficients where ~ c =c Ris Discount max s. t. ~ a = a Ris Discount ~ ~ b =b Ris Discount c ~ A ~ b ~ o E V Maximize E(income)  RAP * Variance(income) Expected utility Maximize Sum(p,Probability(p)*U[Wealth(p)]) S.T. Wealth(p)=InitWealth Income(p) for all p Income(p)=C(p)* for all p Safety First based Maximize Sum(p,Probability(p)*Income(p)) S.T. Income(p)=C(p)* for all p Income(p)$safety for all p B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 7
8 Including Non Adaptive Ris First Ris Model Marowitz meanvariance portfolio choice formulation Given Problem max sum(invest, moneyinvest(invest)*avgreturn(invest)) s.t sum(invest,moneyinvest(invest)*price(invest))# funds Marowitz observed not all money is invested in the highest valued stoc Inconsistent with LP formulation Why? Not a basic solution Marowitz posed the hypothesis that average returns and the variance of returns were important B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 8
9 Including Non Adaptive Ris EV Model Statistical Bacground Given a linear obective function Z = c c Where, are decision variables c, c are uncertain parameters Then Z is distributed with mean and variance Z = σ c c Z = s s s a Defining terms s ii is the variance of obective function coefficient of i, which is calculated using s i = (c i  c i ) /N where s i c i c i is the th observation on the obective value of i and N the number of observations. for i is the covariance of the obective function coefficients between i and, calculated by the formula s i = (c i c i )(c c )/N. Note s i = s i. is the mean value of the obective function coefficient associated with i, calculated by c i = c i /N. (Assuming an equally liely probability of occurrence.) B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 9
10 Including Non Adaptive Ris EV Model Statistical Bacground In matrix terms the mean and variance of Z are Z = c σ Z = where in the two by two case Z = E = [ c c ] s σ Z = s Marowitz Formulation ' S Z [ ] s s Min s. t. σ Z E = K Freund Formulation Min E φ σ Z or Min E  RAP * Variance Why Use Freund reuse of RAP and Transferability B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 0
11 Including Non Adaptive Ris EV Model Commonly Used Formulation Min E  RAP * Variance Max s. t. c φ s funds 0 for all Where E = expected value of risy c times choice of x Var = sum of var time x squared minus twice covariance times x s N = ris aversion parameter B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
12 Including Non Adaptive Ris EV Example Data Data for EV Example  Returns by Stoc and Event Stoc Returns by Stoc and Event Stoc Stoc Stoc Stoc Event Event Event 8 6 Event Event Event6 0 5 Event76 Event Event9 7 5 Event Stoc Stoc Stoc Stoc Price Mean Returns and Variance Parameters for Stoc Example Stoc Stoc Stoc Stoc Mean Returns VarianceCovariance Matrix Stoc Stoc Stoc Stoc Stoc Stoc Stoc Stoc B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
13 Including Non Adaptive Ris EV Model Example Max E  N F = E  RAP * Variance Max s. t. c φ s funds 0 for all or for the example Max s. t. [ ] φ [ ] or Max ϕ s. t B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
14 Including Non Adaptive Ris GAMS Formulation (EVPORTFO.GMS) SETS STOCKS POTENTIAL INVESTMENTS / BUYSTOCK*BUYSTOCK / EVENTS EQUALLY LIKELY RETURN STATES OF NATURE /EVENT*EVENT0 / ; ALIAS (STOCKS,STOCK); PARAMETERS PRICES(STOCKS) PURCHASE PRICES OF THE STOCKS / BUYSTOCK BUYSTOCK 0 BUYSTOCK 8 BUYSTOCK 6 / ; SCALAR FUNDS TOTAL INVESTABLE FUNDS / 500 / ; TABLE RETURNS(EVENTS,STOCKS) RETURNS BY STATE OF NATURE EVENT BUYSTOCK BUYSTOCK BUYSTOCK BUYSTOCK EVENT EVENT EVENT 8 6 EVENT EVENT EVENT6 0 5 EVENT76 EVENT EVENT9 7 5 EVENT PARAMETERS MEAN (STOCKS) MEAN RETURNS TO (STOCKS) COVAR(STOCK,STOCKS) VARIANCE COVARIANCE MATRI; MEAN(STOCKS) = SUM(EVENTS, RETURNS(EVENTS,STOCKS) / CARD(EVENTS) ); COVAR(STOCK,STOCKS)=SUM(EVENTS,(RETURNS(EVENTS,STOCKS)  MEAN(STOCKS)) *(RETURNS(EVENTS,STOCK) MEAN(STOCK)))/CARD(EVENTS); SCALAR RAP RISK AVERSION PARAMETER / 0.0 / ; POSITIVE VARIABLES INVEST(STOCKS) MONEY INVESTED IN EACH STOCK VARIABLE OBJ NUMBER TO BE MAIMIZED ; EQUATIONS OBJJ OBJECTIVE FUNCTION OBJJ.. OBJ =E= INVESTAV INVESTMENT FUNDS AVAILABLE ; SUM(STOCKS, MEAN(STOCKS) * INVEST(STOCKS))  RAP*(SUM(STOCK, SUM(STOCKS, INVEST(STOCK)* COVAR(STOCK,STOCKS) * INVEST(STOCKS)))); INVESTAV.. SUM(STOCKS, PRICES(STOCKS) * INVEST(STOCKS)) =L= FUNDS ; MODEL EVPORTFOL /ALL/ ; SOLVE EVPORTFOL USING NLP MAIMIZING OBJ ; SCALAR VAR THE VARIANCE ; VAR = SUM(STOCK, SUM(STOCKS,INVEST.L(STOCK)*COVAR(STOCK,STOCKS)*INVEST.L(STOCKS))) SET RAPS RISK AVERSION PARAMETERS /R0*R5/ PARAMETER RISKAVER(RAPS) RISK AVERSION COEFICIENT BY RISK AVERSION PARAMETER/ R ,R ,R ,R ,R ,R ,R R ,R ,R ,R ,R 0.050,R ,R R ,R ,R ,R ,R ,R ,R R ,R 5.,R 0.,R 0., R5 80./ PARAMETER OUTPUT(*,RAPS) RESULTS FROM MODEL RUNS WITH VARYING RAP LOOP (RAPS,RAP=RISKAVER(RAPS); SOLVE EVPORTFOL USING NLP MAIMIZING OBJ ; VAR = SUM(STOCK, SUM(STOCKS, INVEST.L(STOCK)* COVAR(STOCK,STOCKS) * INVEST.L(STOCKS))) ; OUTPUT("RAP",RAPS)=RAP; OUTPUT(STOCKS,RAPS)=INVEST.L(STOCKS); OUTPUT("OBJ",RAPS)=OBJ.L; OUTPUT("MEAN",RAPS)=SUM(STOCKS, MEAN(STOCKS) * INVEST.L(STOCKS)); OUTPUT("VAR",RAPS) = VAR;OUTPUT("STD",RAPS)=SQRT(VAR); OUTPUT("SHADPRICE",RAPS)=INVESTAV.M; OUTPUT("IDLE",RAPS)=FUNDSINVESTAV.L ); B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
15 Including Non Adaptive Ris EV Example and Solution (EVPORTFO.GMS) Max [ ] φ [ ] s. t EV Example Solutions for Alternative Ris Aversion Parameters RAP BUYSTOCK BUYSTOCK OBJ MEAN VAR STD SHADPRICE RAP BUYSTOCK BUYSTOCK BUYSTOCK BUYSTOCK.68 OBJ MEAN VAR STD SHADPRICE RAP BUYSTOCK BUYSTOCK BUYSTOCK BUYSTOCK OBJ MEAN VAR STD SHADPRICE IDLE FUNDS 68.0 B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 5
16 Mean Income Including Non Adaptive Ris EV Example and Frontier (EVPORTFO.GMS) Frontier EV Frontier Variance of Income RAP BUYSTOCK BUYSTOCK OBJ MEAN VAR STD SHADPRICE RAP BUYSTOCK BUYSTOCK BUYSTOCK BUYSTOCK.68 OBJ MEAN VAR STD SHADPRICE B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 6
17 Max s. t. r c Including Non Adaptive Ris Alternative EV Model S p S S E E φ( p ( S E) ) funds = 0 for all = 0 0 for all E unrestricted for all α where identifies the stoc possibilities; identifies the states of nature; is amount of stoc bought; r is the cost of buying ; c is the uncertain yield of stoc realized under state of nature when the buying ; S is the income from stocs under state of nature ; p is the probability of state of nature ; E is average income; M is a ris aversion parameter " is for EV models or ½ for E STD models B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 7
18 Max s. t. r c Including Non Adaptive Ris Alternative EV Model S p S S E E E φ( p ( S E) ) α funds = 0 for all = 0 0 for all unrestricted for all Ob Depicts maximization of expected income minus the ris aversion parameter times the variance of income First constraint limits funds available Second equation sums income by state of nature Third weights income by state of nature by probability forming expected income Last two requires investment variables to be nonnegative but allows income to be positive or negative. B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 8
19 Including Non Adaptive Ris Alternative EV Model (EVPORTF.GMS) 0 / ) /0*( = = = = = = = = = = = E I I I I I I I I I I I s t E I E Max R H S P N O E R I I I I a C S G O N N N N v O I A A A W V V V V r e v  r e v  r e g E T I T I C E E E E O r F I J J N S S S S B e F V, V, T T T T T J v S E S E S S OBJJ C = 0 INVESTAV E E E E < F 0 r D D D D = 0 5 e D D E D = 0 5 v D D E D = 0 5 e D D D = 0 5 n 5 D D E D = 0 5 u 6 D E E D = 0 5 e 7 D E D = D D E D = 0 5 r 9 D D E D = 0 5 e 0 D D D = 0 5 avrevenue A A A A A A A A A A = 0 0 B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 9
20 Including Non Adaptive Ris Alternative EV Model (EVPORTF.GMS) SETS STOCKS POTENTIAL INVESTMENTS / BUYSTOCK*BUYSTOCK / EVENTS EQUALLY LIKELY STATES OF NATURE /EVENT*EVENT0 / ;... SCALAR RAP RISK AVERSION PARAMETER / 0.0 / alpha term if is var if 0.5 is std err // n sample size; n=card(events); POSITIVE VARIABLES INVEST(STOCKS) MONEY INVESTED IN EACH STOCK VARIABLE OBJ NUMBER TO BE MAIMIZED rev(events) revenue by event avgrev average revenue; EQUATIONS OBJJ OBJECTIVE FUNCTION INVESTAV INVESTMENT FUNDS AVAILABLE revenue(events) accounts revenue by event avrevenue accounts average revenue ; OBJJ.. OBJ =E= avgrev  RAP* (SUM(events,/n*(rev(events)avgrev)*(rev(events)avgrev)))**alpha; INVESTAV.. SUM(STOCKS, PRICES(STOCKS) * INVEST(STOCKS)) =L= FUNDS ; revenue(events).. sum(stocs,returns(events,stocks)*invest(stocks))=e=rev(events); avrevenue.. sum(events,/n*rev(events))=e=avgrev; MODEL EVPORTFOL /ALL/ ; SOLVE EVPORTFOL USING NLP MAIMIZING OBJ ; SCALAR VAR THE VARIANCE ; VAR = (SUM(events, /n*(rev.l(events)avgrev.l)*(rev.l(events)avgrev.l))) ; SET RAPS RISK AVERSION PARAMETERS /R0*R5/ PARAMETER RISKAVER(RAPS) RISK AVERSION COEFICIENT BY RISK AVERSION / R ,R ,R ,R ,R ,R ,R R ,R ,R ,R ,R 0.050,R ,R R ,R ,R ,R ,R ,R ,R R ,R 5.,R 0.,R 0., R5 80./ PARAMETER OUTPUT(*,RAPS) RESULTS FROM MODEL RUNS WITH VARYING RAP LOOP (RAPS,RAP=RISKAVER(RAPS); SOLVE EVPORTFOL USING NLP MAIMIZING OBJ ; var=(sum(events,/n*(rev.l(events)avgrev.l)*(rev.l(events)avgrev.l))) ; OUTPUT("RAP",RAPS)=RAP; OUTPUT(STOCKS,RAPS)=INVEST.L(STOCKS); OUTPUT("OBJ",RAPS)=OBJ.L; OUTPUT("MEAN",RAPS)=avgrev.l; OUTPUT("VAR",RAPS) = VAR; OUTPUT("STD",RAPS)=SQRT(VAR); OUTPUT("SHADPRICE",RAPS)=INVESTAV.M; OUTPUT("IDLE",RAPS)=FUNDSINVESTAV.L ); parameter graphit (*,raps,*); graphit("frontier",raps,"mean")=output("mean",raps); graphit("frontier",raps,"var")=output("std",raps)**; alpha=0.5; LOOP (RAPS,RAP=RISKAVER(RAPS); SOLVE EVPORTFOL USING NLP MAIMIZING OBJ ; var=(sum(events,/n*(rev.l(events)avgrev.l)*(rev.l(events)avgrev.l))) ; OUTPUT("RAP",RAPS)=RAP; OUTPUT(STOCKS,RAPS)=INVEST.L(STOCKS); OUTPUT("OBJ",RAPS)=OBJ.L; OUTPUT("MEAN",RAPS)=avgrev.l; OUTPUT("VAR",RAPS) = VAR; OUTPUT("STD",RAPS)=SQRT(VAR); OUTPUT("SHADPRICE",RAPS)=INVESTAV.M; OUTPUT("IDLE",RAPS)=FUNDSINVESTAV.L ); DISPLAY OUTPUT; B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 0
21 Including Non Adaptive Ris Alternative EV Model (EVPORTF.GMS) Mean Income Frontier Std Err EV Frontier Variance of Income EV solutions are identical with earlier, E Standard Error is close R0 R R R R R5 BUYSTOCK BUYSTOCK RAP E E E OBJ MEAN VAR STD SHADPRICE R6 R7 R8 R9 R0 R BUYSTOCK BUYSTOCK RAP OBJ MEAN VAR STD SHADPRICE B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
22 Including Non Adaptive Ris Dissecting the GAMS formulation Graphing parameter graphit (*,raps,*); graphit("frontier",raps,"mean")=output("mean",raps); graphit("frontier",raps,"var")=output("std",raps)**; *$include gnu_opt.gms * titles... more solves... graphit("std Err",raps,"Mean")=OUTPUT("MEAN",RAPS); graphit("std Err",raps,"Var")=OUTPUT("std",RAPS)**; $setglobal gp_title "EV Frontier " $setglobal gp_xlabel "Variance of Income" $setglobal gp_ylabel "Mean Income" $batinclude gnupltxy graphit mean var This is done using a GNUPLOT interface originally developed by Rutherford but modified to gnupltxy as documented on the Web page agecon.tamu.edu/faculty/mccarl B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
23 Including Non Adaptive Ris Modeling Support from GAMSCHK Nonlinear Models ## rev('event') SOLUTION VALUE.697 EQN Ai Ui Ai*Ui OBJJ ***0.706E E0 revenue('event') avrevenue TRUE REDUCED COST E ## EQU OBJJ ## OBJJ VAR Ai Ai* OBJ rev('event') ***0.706E rev('event') *** E rev('event') *** rev('event') *** E E0 rev('event5') *** rev('event6') *** rev('event7') *** E rev('event8') *** rev('event9') ***0.56E rev('event0') *** avgrev *** =E= =E= RHS COEFF E00 *** mars nonl;inear terms Starting point and accuracy is an issue B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
24 Including Firm Level Ris Finding a Ris Aversion Parameter EMF is one tailed confidence interval when M= under normality confidence interval is 8% M=.96 interval is 97.5% Also EV relation Max s. t. c ψσ A ( ) b 0 versus Max s. t. c θσ ( ) A b 0 σ ( ) σ ( ) c ψσ ( ) λa= 0 c θ λa = 0 which equates when θ ψ = σ ( ) Given we generally find θ in a range between 0 and 5 this implies 5 0 ψ σ ( ) B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page
25 Including Firm Level Ris Forming Probability Distributions Probability distributions state the relative frequency of occurrence of a set of mutually exclusive events. Finding Probability Distributions Based on Obective Data Desirable Characteristics ) each of the states of nature must be mutually exclusive; ) probability of occurrence of each of the states of nature must be an unbiased measure of the current probability of that state of nature occurring; ) the sum of the probabilities across the states of nature must equal one Second property is the most troubling in when using obective, historical data. trends, events B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 5
26 Including Firm Level Ris B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 6
27 Including Firm Level Ris 0.5 Figure 5. Probability Distribution 0. Probability Stationary Price Historic Price Real Price Price B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 7
28 Including Firm Level Ris General Lessons Learned on Obective Probabilities Use obective data trends and other systematic effects can bias Use a procedure lie regression to develop values expected One may find residual terms are hetereosedastic B.A. McCarl, May 00 Including Ris Part Non Adaptive (firmris) page 8
Agricultural Economics 622, Spring 2004 Midterm Exam on Math programming Topic
Agricultural Economics 622, Spring 2004 Midterm Exam on Math programming Topic 1. (25 points) Temple Outland forest group is trying to set up a plan for utilizing its land and wood processing facilities.
More informationSolving Linear Programs in Excel
Notes for AGEC 622 Bruce McCarl Regents Professor of Agricultural Economics Texas A&M University Thanks to Michael Lau for his efforts to prepare the earlier copies of this. 1 http://ageco.tamu.edu/faculty/mccarl/622class/
More informationOPTIMAL CHOICE UNDER SHORT SELL LIMIT WITH SHARPE RATIO AS CRITERION AMONG MULTIPLE ASSETS
OPTIMAL CHOICE UNDER SHORT SELL LIMIT WITH SHARPE RATIO AS CRITERION AMONG MULTIPLE ASSETS Ruoun HUANG *, Yiran SHENG ** Abstract: This article is the term paper of the course Investments. We mainly focus
More informationProblem Set 6  Solutions
ECO573 Financial Economics Problem Set 6  Solutions 1. Debt Restructuring CAPM. a Before refinancing the stoc the asset have the same beta: β a = β e = 1.2. After restructuring the company has the same
More informationPEST  Beyond Basic Model Calibration. Presented by Jon Traum
PEST  Beyond Basic Model Calibration Presented by Jon Traum Purpose of Presentation Present advance techniques available in PEST for model calibration High level overview Inspire more people to use PEST!
More informationDecision making under uncertainty  is sensitivity analysis of any use?
Decision making under uncertainty  is sensitivity analysis of any use? Stein W. Wallace Molde University College 1 Decision making under uncertainty Sensitivity analysis Parametric optimization Whatif
More informationA LogRobust Optimization Approach to Portfolio Management
A LogRobust Optimization Approach to Portfolio Management Dr. Aurélie Thiele Lehigh University Joint work with Ban Kawas Research partially supported by the National Science Foundation Grant CMMI0757983
More informationTesting for Granger causality between stock prices and economic growth
MPRA Munich Personal RePEc Archive Testing for Granger causality between stock prices and economic growth Pasquale Foresti 2006 Online at http://mpra.ub.unimuenchen.de/2962/ MPRA Paper No. 2962, posted
More information5. Linear Regression
5. Linear Regression Outline.................................................................... 2 Simple linear regression 3 Linear model............................................................. 4
More informationSensex Realized Volatility Index
Sensex Realized Volatility Index Introduction: Volatility modelling has traditionally relied on complex econometric procedures in order to accommodate the inherent latent character of volatility. Realized
More informationNCSS Statistical Software Principal Components Regression. In ordinary least squares, the regression coefficients are estimated using the formula ( )
Chapter 340 Principal Components Regression Introduction is a technique for analyzing multiple regression data that suffer from multicollinearity. When multicollinearity occurs, least squares estimates
More informationA Review of Cross Sectional Regression for Financial Data You should already know this material from previous study
A Review of Cross Sectional Regression for Financial Data You should already know this material from previous study But I will offer a review, with a focus on issues which arise in finance 1 TYPES OF FINANCIAL
More informationReview for Exam 2. Instructions: Please read carefully
Review for Exam 2 Instructions: Please read carefully The exam will have 25 multiple choice questions and 5 work problems You are not responsible for any topics that are not covered in the lecture note
More informationPlease follow the directions once you locate the Stata software in your computer. Room 114 (Business Lab) has computers with Stata software
STATA Tutorial Professor Erdinç Please follow the directions once you locate the Stata software in your computer. Room 114 (Business Lab) has computers with Stata software 1.Wald Test Wald Test is used
More informationPrinciples of demand management Airline yield management Determining the booking limits. » A simple problem» Stochastic gradients for general problems
Demand Management Principles of demand management Airline yield management Determining the booking limits» A simple problem» Stochastic gradients for general problems Principles of demand management Issues:»
More informationDr Christine Brown University of Melbourne
Enhancing Risk Management and Governance in the Region s Banking System to Implement Basel II and to Meet Contemporary Risks and Challenges Arising from the Global Banking System Training Program ~ 8 12
More informationconstraint. Let us penalize ourselves for making the constraint too big. We end up with a
Chapter 4 Constrained Optimization 4.1 Equality Constraints (Lagrangians) Suppose we have a problem: Maximize 5, (x 1, 2) 2, 2(x 2, 1) 2 subject to x 1 +4x 2 =3 If we ignore the constraint, we get the
More informationThe Capital Asset Pricing Model
Finance 400 A. Penati  G. Pennacchi The Capital Asset Pricing Model Let us revisit the problem of an investor who maximizes expected utility that depends only on the expected return and variance (or standard
More informationMultiple Linear Regression in Data Mining
Multiple Linear Regression in Data Mining Contents 2.1. A Review of Multiple Linear Regression 2.2. Illustration of the Regression Process 2.3. Subset Selection in Linear Regression 1 2 Chap. 2 Multiple
More informationSimplex Method. Introduction:
Introduction: In the previous chapter, we discussed about the graphical method for solving linear programming problems. Although the graphical method is an invaluable aid to understand the properties of
More informationForecasting in supply chains
1 Forecasting in supply chains Role of demand forecasting Effective transportation system or supply chain design is predicated on the availability of accurate inputs to the modeling process. One of the
More informationChapter 8: Interval Estimates and Hypothesis Testing
Chapter 8: Interval Estimates and Hypothesis Testing Chapter 8 Outline Clint s Assignment: Taking Stock Estimate Reliability: Interval Estimate Question o Normal Distribution versus the Student tdistribution:
More informationSimple Linear Regression Inference
Simple Linear Regression Inference 1 Inference requirements The Normality assumption of the stochastic term e is needed for inference even if it is not a OLS requirement. Therefore we have: Interpretation
More informationMISSING DATA TECHNIQUES WITH SAS. IDRE Statistical Consulting Group
MISSING DATA TECHNIQUES WITH SAS IDRE Statistical Consulting Group ROAD MAP FOR TODAY To discuss: 1. Commonly used techniques for handling missing data, focusing on multiple imputation 2. Issues that could
More informationGeostatistics Exploratory Analysis
Instituto Superior de Estatística e Gestão de Informação Universidade Nova de Lisboa Master of Science in Geospatial Technologies Geostatistics Exploratory Analysis Carlos Alberto Felgueiras cfelgueiras@isegi.unl.pt
More informationChapter 2 Portfolio Management and the Capital Asset Pricing Model
Chapter 2 Portfolio Management and the Capital Asset Pricing Model In this chapter, we explore the issue of risk management in a portfolio of assets. The main issue is how to balance a portfolio, that
More informationPaper No 19. FINALTERM EXAMINATION Fall 2009 MTH302 Business Mathematics & Statistics (Session  2) Ref No: Time: 120 min Marks: 80
Paper No 19 FINALTERM EXAMINATION Fall 2009 MTH302 Business Mathematics & Statistics (Session  2) Ref No: Time: 120 min Marks: 80 Question No: 1 ( Marks: 1 )  Please choose one Scatterplots are used
More informationAgenda. Managing Uncertainty in the Supply Chain. The Economic Order Quantity. Classic inventory theory
Agenda Managing Uncertainty in the Supply Chain TIØ485 Produkjons og nettverksøkonomi Lecture 3 Classic Inventory models Economic Order Quantity (aka Economic Lot Size) The (s,s) Inventory Policy Managing
More informationUsing Microsoft Excel to build Efficient Frontiers via the Mean Variance Optimization Method
Using Microsoft Excel to build Efficient Frontiers via the Mean Variance Optimization Method Submitted by John Alexander McNair ID #: 0061216 Date: April 14, 2003 The Optimal Portfolio Problem Consider
More informationVariance of OLS Estimators and Hypothesis Testing. Randomness in the model. GM assumptions. Notes. Notes. Notes. Charlie Gibbons ARE 212.
Variance of OLS Estimators and Hypothesis Testing Charlie Gibbons ARE 212 Spring 2011 Randomness in the model Considering the model what is random? Y = X β + ɛ, β is a parameter and not random, X may be
More informationIs the Forward Exchange Rate a Useful Indicator of the Future Exchange Rate?
Is the Forward Exchange Rate a Useful Indicator of the Future Exchange Rate? Emily Polito, Trinity College In the past two decades, there have been many empirical studies both in support of and opposing
More informationChapter 5 Risk and Return ANSWERS TO SELECTED ENDOFCHAPTER QUESTIONS
Chapter 5 Risk and Return ANSWERS TO SELECTED ENDOFCHAPTER QUESTIONS 51 a. Standalone risk is only a part of total risk and pertains to the risk an investor takes by holding only one asset. Risk is
More informationThe Effects of Start Prices on the Performance of the Certainty Equivalent Pricing Policy
BMI Paper The Effects of Start Prices on the Performance of the Certainty Equivalent Pricing Policy Faculty of Sciences VU University Amsterdam De Boelelaan 1081 1081 HV Amsterdam Netherlands Author: R.D.R.
More information1 Portfolio mean and variance
Copyright c 2005 by Karl Sigman Portfolio mean and variance Here we study the performance of a oneperiod investment X 0 > 0 (dollars) shared among several different assets. Our criterion for measuring
More informationPortfolio Construction in a Regression Framework
Portfolio Construction in a Regression Framework James Sefton June 2007 Motivation The approach can be used to answer the following questions: Are the weights in my portfolio the result of one or two abnormal
More informationCorporate Defaults and Large Macroeconomic Shocks
Corporate Defaults and Large Macroeconomic Shocks Mathias Drehmann Bank of England Andrew Patton London School of Economics and Bank of England Steffen Sorensen Bank of England The presentation expresses
More informationDemand Forecasting LEARNING OBJECTIVES IEEM 517. 1. Understand commonly used forecasting techniques. 2. Learn to evaluate forecasts
IEEM 57 Demand Forecasting LEARNING OBJECTIVES. Understand commonly used forecasting techniques. Learn to evaluate forecasts 3. Learn to choose appropriate forecasting techniques CONTENTS Motivation Forecast
More information11. OVERVIEW OF THE INVESTMENT PORTFOLIO SOFTWARE
11. OVERVIEW OF THE INVESTMENT PORTFOLIO SOFTWARE The Investment Portfolio software was developed by Edwin J. Elton, Martin J. Gruber and Christopher R. Blake, in conjunction with IntelliPro, Inc., to
More information2. Efficiency and Perfect Competition
General Equilibrium Theory 1 Overview 1. General Equilibrium Analysis I Partial Equilibrium Bias 2. Efficiency and Perfect Competition 3. General Equilibrium Analysis II The Efficiency if Competition The
More informationproblem arises when only a nonrandom sample is available differs from censored regression model in that x i is also unobserved
4 Data Issues 4.1 Truncated Regression population model y i = x i β + ε i, ε i N(0, σ 2 ) given a random sample, {y i, x i } N i=1, then OLS is consistent and efficient problem arises when only a nonrandom
More informationOLS is not only unbiased it is also the most precise (efficient) unbiased estimation technique  ie the estimator has the smallest variance
Lecture 5: Hypothesis Testing What we know now: OLS is not only unbiased it is also the most precise (efficient) unbiased estimation technique  ie the estimator has the smallest variance (if the GaussMarkov
More informationEXERCISES FROM HULL S BOOK
EXERCISES FROM HULL S BOOK 1. Three put options on a stock have the same expiration date, and strike prices of $55, $60, and $65. The market price are $3, $5, and $8, respectively. Explain how a butter
More informationt = 1 2 3 1. Calculate the implied interest rates and graph the term structure of interest rates. t = 1 2 3 X t = 100 100 100 t = 1 2 3
MØA 155 PROBLEM SET: Summarizing Exercise 1. Present Value [3] You are given the following prices P t today for receiving risk free payments t periods from now. t = 1 2 3 P t = 0.95 0.9 0.85 1. Calculate
More informationECON 142 SKETCH OF SOLUTIONS FOR APPLIED EXERCISE #2
University of California, Berkeley Prof. Ken Chay Department of Economics Fall Semester, 005 ECON 14 SKETCH OF SOLUTIONS FOR APPLIED EXERCISE # Question 1: a. Below are the scatter plots of hourly wages
More informationBusiness Statistics. Successful completion of Introductory and/or Intermediate Algebra courses is recommended before taking Business Statistics.
Business Course Text Bowerman, Bruce L., Richard T. O'Connell, J. B. Orris, and Dawn C. Porter. Essentials of Business, 2nd edition, McGrawHill/Irwin, 2008, ISBN: 9780073319889. Required Computing
More informationChapter 6: Multivariate Cointegration Analysis
Chapter 6: Multivariate Cointegration Analysis 1 Contents: Lehrstuhl für Department Empirische of Wirtschaftsforschung Empirical Research and und Econometrics Ökonometrie VI. Multivariate Cointegration
More informationFrom the help desk: Swamy s randomcoefficients model
The Stata Journal (2003) 3, Number 3, pp. 302 308 From the help desk: Swamy s randomcoefficients model Brian P. Poi Stata Corporation Abstract. This article discusses the Swamy (1970) randomcoefficients
More informationSection A. Index. Section A. Planning, Budgeting and Forecasting Section A.2 Forecasting techniques... 1. Page 1 of 11. EduPristine CMA  Part I
Index Section A. Planning, Budgeting and Forecasting Section A.2 Forecasting techniques... 1 EduPristine CMA  Part I Page 1 of 11 Section A. Planning, Budgeting and Forecasting Section A.2 Forecasting
More informationWooldridge, Introductory Econometrics, 4th ed. Multiple regression analysis:
Wooldridge, Introductory Econometrics, 4th ed. Chapter 4: Inference Multiple regression analysis: We have discussed the conditions under which OLS estimators are unbiased, and derived the variances of
More informationCoefficient of Determination
Coefficient of Determination The coefficient of determination R 2 (or sometimes r 2 ) is another measure of how well the least squares equation ŷ = b 0 + b 1 x performs as a predictor of y. R 2 is computed
More informationChoice under Uncertainty
Choice under Uncertainty Part 1: Expected Utility Function, Attitudes towards Risk, Demand for Insurance Slide 1 Choice under Uncertainty We ll analyze the underlying assumptions of expected utility theory
More informationSummary of specified general model for CHP system
Fakulteta za Elektrotehniko Eva Thorin, Heike Brand, Christoph Weber Summary of specified general model for CHP system OSCOGEN Deliverable D1.4 Contract No. ENK5CT200000094 Project cofunded by the
More informationSensitivity Analysis with Excel
Sensitivity Analysis with Excel 1 Lecture Outline Sensitivity Analysis Effects on the Objective Function Value (OFV): Changing the Values of Decision Variables Looking at the Variation in OFV: Excel One
More informationThe Big Picture. Correlation. Scatter Plots. Data
The Big Picture Correlation Bret Hanlon and Bret Larget Department of Statistics Universit of Wisconsin Madison December 6, We have just completed a length series of lectures on ANOVA where we considered
More informationBlackLitterman Return Forecasts in. Tom Idzorek and Jill Adrogue Zephyr Associates, Inc. September 9, 2003
BlackLitterman Return Forecasts in Tom Idzorek and Jill Adrogue Zephyr Associates, Inc. September 9, 2003 Using BlackLitterman Return Forecasts for Asset Allocation Results in Diversified Portfolios
More informationOPTIMIZATION AND FORECASTING WITH FINANCIAL TIME SERIES
OPTIMIZATION AND FORECASTING WITH FINANCIAL TIME SERIES Allan Din Geneva Research Collaboration Notes from seminar at CERN, June 25, 2002 General scope of GRC research activities Econophysics paradigm
More informationPorter, White & Company
Porter, White & Company Optimizing the Fixed Income Component of a Portfolio White Paper, September 2009, Number IM 17.2 In the White Paper, Comparison of Fixed Income Fund Performance, we show that a
More informationHolding Period Return. Return, Risk, and Risk Aversion. Percentage Return or Dollar Return? An Example. Percentage Return or Dollar Return? 10% or 10?
Return, Risk, and Risk Aversion Holding Period Return Ending Price  Beginning Price + Intermediate Income Return = Beginning Price R P t+ t+ = Pt + Dt P t An Example You bought IBM stock at $40 last month.
More informationOverview of Violations of the Basic Assumptions in the Classical Normal Linear Regression Model
Overview of Violations of the Basic Assumptions in the Classical Normal Linear Regression Model 1 September 004 A. Introduction and assumptions The classical normal linear regression model can be written
More informationA Branch and Bound Algorithm for Solving the Binary Bilevel Linear Programming Problem
A Branch and Bound Algorithm for Solving the Binary Bilevel Linear Programming Problem John Karlof and Peter Hocking Mathematics and Statistics Department University of North Carolina Wilmington Wilmington,
More information2. Filling Data Gaps, Data validation & Descriptive Statistics
2. Filling Data Gaps, Data validation & Descriptive Statistics Dr. Prasad Modak Background Data collected from field may suffer from these problems Data may contain gaps ( = no readings during this period)
More informationFinancial Market Microstructure Theory
The Microstructure of Financial Markets, de Jong and Rindi (2009) Financial Market Microstructure Theory Based on de Jong and Rindi, Chapters 2 5 Frank de Jong Tilburg University 1 Determinants of the
More informationOption Portfolio Modeling
Value of Option (Total=Intrinsic+Time Euro) Option Portfolio Modeling Harry van Breen www.besttheindex.com Email: h.j.vanbreen@besttheindex.com Introduction The goal of this white paper is to provide
More informationGraduate Macro Theory II: The Real Business Cycle Model
Graduate Macro Theory II: The Real Business Cycle Model Eric Sims University of Notre Dame Spring 2011 1 Introduction This note describes the canonical real business cycle model. A couple of classic references
More informationStatistics in Retail Finance. Chapter 6: Behavioural models
Statistics in Retail Finance 1 Overview > So far we have focussed mainly on application scorecards. In this chapter we shall look at behavioural models. We shall cover the following topics: Behavioural
More informationChapter 2: Binomial Methods and the BlackScholes Formula
Chapter 2: Binomial Methods and the BlackScholes Formula 2.1 Binomial Trees We consider a financial market consisting of a bond B t = B(t), a stock S t = S(t), and a calloption C t = C(t), where the
More informationFinancial Markets. Itay Goldstein. Wharton School, University of Pennsylvania
Financial Markets Itay Goldstein Wharton School, University of Pennsylvania 1 Trading and Price Formation This line of the literature analyzes the formation of prices in financial markets in a setting
More informationMATH4427 Notebook 2 Spring 2016. 2 MATH4427 Notebook 2 3. 2.1 Definitions and Examples... 3. 2.2 Performance Measures for Estimators...
MATH4427 Notebook 2 Spring 2016 prepared by Professor Jenny Baglivo c Copyright 20092016 by Jenny A. Baglivo. All Rights Reserved. Contents 2 MATH4427 Notebook 2 3 2.1 Definitions and Examples...................................
More informationFactors Influencing Price/Earnings Multiple
Learning Objectives Foundation of Research Forecasting Methods Factors Influencing Price/Earnings Multiple Passive & Active Asset Management Investment in Foreign Markets Introduction In the investment
More informationMODERN PORTFOLIO THEORY AND INVESTMENT ANALYSIS
MODERN PORTFOLIO THEORY AND INVESTMENT ANALYSIS EIGHTH EDITION INTERNATIONAL STUDENT VERSION EDWIN J. ELTON Leonard N. Stern School of Business New York University MARTIN J. GRUBER Leonard N. Stern School
More informationJava Modules for Time Series Analysis
Java Modules for Time Series Analysis Agenda Clustering Nonnormal distributions Multifactor modeling Implied ratings Time series prediction 1. Clustering + Cluster 1 Synthetic Clustering + Time series
More informationIntegrated MultiEchelon Supply Chain Design with Inventories under Uncertainty: MINLP Models, Computational Strategies
Integrated MultiEchelon Supply Chain Design with Inventories under Uncertainty: MINLP Models, Computational Strategies Fengqi You, Ignacio E. Grossmann * Department of Chemical Engineering, Carnegie Mellon
More information9 Hedging the Risk of an Energy Futures Portfolio UNCORRECTED PROOFS. Carol Alexander 9.1 MAPPING PORTFOLIOS TO CONSTANT MATURITY FUTURES 12 T 1)
Helyette Geman c0.tex V  0//0 :00 P.M. Page Hedging the Risk of an Energy Futures Portfolio Carol Alexander This chapter considers a hedging problem for a trader in futures on crude oil, heating oil and
More informationBivariate Regression Analysis. The beginning of many types of regression
Bivariate Regression Analysis The beginning of many types of regression TOPICS Beyond Correlation Forecasting Two points to estimate the slope Meeting the BLUE criterion The OLS method Purpose of Regression
More informationUsing least squares Monte Carlo for capital calculation 21 November 2011
Life Conference and Exhibition 2011 Adam Koursaris, Peter Murphy Using least squares Monte Carlo for capital calculation 21 November 2011 Agenda SCR calculation Nested stochastic problem Limitations of
More informationLecture 2: Delineating efficient portfolios, the shape of the meanvariance frontier, techniques for calculating the efficient frontier
Lecture 2: Delineating efficient portfolios, the shape of the meanvariance frontier, techniques for calculating the efficient frontier Prof. Massimo Guidolin Portfolio Management Spring 2016 Overview The
More informationC&O 370 Deterministic OR Models Winter 2011
C&O 370 Deterministic OR Models Winter 2011 Assignment 1 Due date: Friday Jan. 21, 2011 Assignments are due at the start of class on the due date. Write your name and ID# clearly, and underline your last
More informationRisk Budgeting: Concept, Interpretation and Applications
Risk Budgeting: Concept, Interpretation and Applications Northfield Research Conference 005 Eddie Qian, PhD, CFA Senior Portfolio Manager 60 Franklin Street Boston, MA 00 (67) 439637 7538 8//005 The Concept
More informationChicago Booth BUSINESS STATISTICS 41000 Final Exam Fall 2011
Chicago Booth BUSINESS STATISTICS 41000 Final Exam Fall 2011 Name: Section: I pledge my honor that I have not violated the Honor Code Signature: This exam has 34 pages. You have 3 hours to complete this
More informationHandling missing data in Stata a whirlwind tour
Handling missing data in Stata a whirlwind tour 2012 Italian Stata Users Group Meeting Jonathan Bartlett www.missingdata.org.uk 20th September 2012 1/55 Outline The problem of missing data and a principled
More informationThe primary goal of this thesis was to understand how the spatial dependence of
5 General discussion 5.1 Introduction The primary goal of this thesis was to understand how the spatial dependence of consumer attitudes can be modeled, what additional benefits the recovering of spatial
More informationMoral Hazard. Itay Goldstein. Wharton School, University of Pennsylvania
Moral Hazard Itay Goldstein Wharton School, University of Pennsylvania 1 PrincipalAgent Problem Basic problem in corporate finance: separation of ownership and control: o The owners of the firm are typically
More informationEcon 371 Problem Set #3 Answer Sheet
Econ 371 Problem Set #3 Answer Sheet 4.1 In this question, you are told that a OLS regression analysis of third grade test scores as a function of class size yields the following estimated model. T estscore
More information1. What is the critical value for this 95% confidence interval? CV = z.025 = invnorm(0.025) = 1.96
1 Final Review 2 Review 2.1 CI 1propZint Scenario 1 A TV manufacturer claims in its warranty brochure that in the past not more than 10 percent of its TV sets needed any repair during the first two years
More informationMarket Risk: FROM VALUE AT RISK TO STRESS TESTING. Agenda. Agenda (Cont.) Traditional Measures of Market Risk
Market Risk: FROM VALUE AT RISK TO STRESS TESTING Agenda The Notional Amount Approach Price Sensitivity Measure for Derivatives Weakness of the Greek Measure Define Value at Risk 1 Day to VaR to 10 Day
More informationLecture 1: Linear Programming Models. Readings: Chapter 1; Chapter 2, Sections 1&2
Lecture 1: Linear Programming Models Readings: Chapter 1; Chapter 2, Sections 1&2 1 Optimization Problems Managers, planners, scientists, etc., are repeatedly faced with complex and dynamic systems which
More informationMulticollinearity in Regression Models
00 Jeeshim and KUCC65. (0030509) Multicollinearity.doc Introduction Multicollinearity in Regression Models Multicollinearity is a high degree of correlation (linear dependency) among several independent
More informationAPPLICATION OF THE VARMA MODEL FOR SALES FORECAST: CASE OF URMIA GRAY CEMENT FACTORY
APPLICATION OF THE VARMA MODEL FOR SALES FORECAST: CASE OF URMIA GRAY CEMENT FACTORY DOI: 10.2478/tjeb20140005 Ramin Bashir KHODAPARASTI 1 Samad MOSLEHI 2 To forecast sales as reliably as possible is
More informationChapter 1 Introduction to Econometrics
Chapter 1 Introduction to Econometrics Econometrics deals with the measurement of economic relationships. It is an integration of economics, mathematical economics and statistics with an objective to provide
More informationFinance 400 A. Penati  G. Pennacchi Market MicroStructure: Notes on the Kyle Model
Finance 400 A. Penati  G. Pennacchi Market MicroStructure: Notes on the Kyle Model These notes consider the singleperiod model in Kyle (1985) Continuous Auctions and Insider Trading, Econometrica 15,
More informationQuantitative Methods for Finance
Quantitative Methods for Finance Module 1: The Time Value of Money 1 Learning how to interpret interest rates as required rates of return, discount rates, or opportunity costs. 2 Learning how to explain
More informationLinear Programming: Using the Excel Solver
Outline: Linear Programming: Using the Excel Solver We will use Microsoft Excel Solver to solve the four LP examples discussed in last class. 1. The Product Mix Example The Outdoor Furniture Corporation
More informationCHAPTER 5. Exercise Solutions
CHAPTER 5 Exercise Solutions 91 Chapter 5, Exercise Solutions, Principles of Econometrics, e 9 EXERCISE 5.1 (a) y = 1, x =, x = x * * i x i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 y * i (b) (c) yx = 1, x = 16, yx
More informationCHAPTER 9: SERIAL CORRELATION
Serial correlation (or autocorrelation) is the violation of Assumption 4 (observations of the error term are uncorrelated with each other). Pure Serial Correlation This type of correlation tends to be
More informationChapter 2 Introduction to Linear Programming
Chapter 2 Introduction to Linear Programming The key takeaways for the reader from this chapter are listed below: A good understanding of linear programming (LP) problems Formulation of the twovariable
More informationDepartment of Industrial Engineering
Department of Industrial Engineering Master of Engineering Program in Industrial Engineering (International Program) M.Eng. (Industrial Engineering) Plan A Option 2: Total credits required: minimum 39
More informationStatistics 104: Section 6!
Page 1 Statistics 104: Section 6! TF: Deirdre (say: Deardra) Bloome Email: dbloome@fas.harvard.edu Section Times Thursday 2pm3pm in SC 109, Thursday 5pm6pm in SC 705 Office Hours: Thursday 6pm7pm SC
More informationChapter 6 The Tradeoff Between Risk and Return
Chapter 6 The Tradeoff Between Risk and Return MULTIPLE CHOICE 1. Which of the following is an example of systematic risk? a. IBM posts lower than expected earnings. b. Intel announces record earnings.
More informationCombining decision analysis and portfolio management to improve project selection in the exploration and production firm
Journal of Petroleum Science and Engineering 44 (2004) 55 65 www.elsevier.com/locate/petrol Combining decision analysis and portfolio management to improve project selection in the exploration and production
More informationFinal Exam Practice Problem Answers
Final Exam Practice Problem Answers The following data set consists of data gathered from 77 popular breakfast cereals. The variables in the data set are as follows: Brand: The brand name of the cereal
More information