International Journal of Advanced Technology & Engineering Research (IJATER) 1 st International e-conference on Emerging Trends in Technology AN ANLYTICAL APPROACH FOR FAST MOVINGCONSUMER GOODS (FMCG) SECTOR IN INDIAN STOCKMARKET USING DATA MINING TECHNIQUES Kaushar R Ghanchi Head of the Department-BCA Navgujarat College of Computer Applications Ahmedabad. Gujarat. kaush18@gmail.com Ajay D Shah Director Navgujarat College of Computer Applications Ahmedabad. Gujarat. director@navgujaratbca.com ABSTRACT This paper basically deals with identifying frequent patterns from large amount of stock data of FMCG (Fast Moving Consumer Goods) Sector in Indian Capital Market. These frequent patterns are identified based on rise and fall of stock prices of companies of FMCG sector in Indian a Capital Market. We have two stages, in first stage we categorize the FMCG sector s stock data based on Negative growth, Zero growth and Positive growth using Apriori algorithm. In second stage we use Apriori Algorithm to generate useful trends about the behavior of the prices of stock in Capital market. The trend holds to interpret the present and predict the next stock price. Some item-set from sales data indicate market needs and can be used in forecasting which has great potential for decision making, market competition and strategic planning. The objective in this research is to identify or to predict the stock market from the viewpoint of investors. So the investors can invest their shares in the appropriate companies based on Negative growth, Zero growth and Positive growth. These two stage mining process that is Apriori algorithm can generate more useful item-set according to our analysis. INTRODUCTION We are living in the age of Knowledge is a popular saying; however, we are actually living in the data age. Terabytes of data pour into our computer networks, the world wild web(www), and various data storage devices every day from business, society, science and engineering and almost every other aspect of daily life. This explosive growth of available data volume is a result of the computerization of our society. Data mining and knowledge discovery in databases have been attracting a significant amount of research, industry, and media attention of late. What is all the excitement about? Data mining is primarily used today by companies with a strong consumer focus - retail, financial, communication, and marketing organizations. It enables these companies to determine relationships among "internal" factors such as price, product positioning, or staff skills, and "external" factors such as economic indicators, competition, and customer demographics. And, it enables them to determine the impact on sales, customer satisfaction, and corporate profits. Finally, it enables them to "drill down" into summary information to view detail transactional data Mining is an analytic process designed to explore data and in search of consistent patterns and/or systematic relationships between variables, and then to validate the findings by applying the detected patterns to new subsets of data. Stock market is a place where buying and selling of stocks/shares takes place. This trading can be done manually on trading floor or online. Stock Prices are considered to be very dynamic and susceptible to quick changes because of the underlying nature of the financial domain and in part because of the mix of known parameters like High, Low, close prices, P/E Ratio etc and unknown factors (like Election Results, Rumors etc).in this research we have taken the original data sets of Bombay Stock Exchange (BSE) of different companies of FMCG sector such as ISSN No: 2250-3536 E-ICETT 2013 14
Colgate-Palmolive (India) Ltd., Godrej Consumer Products Ltd., Hindustan Unilever Ltd. etc from BSE website and try to classify the FMCG companies in positive Growth, Zero Growth and Negative Growth from the different data fields available from BSE. In this research we have used data mining techniques as clustering algorithm K means to classify the data and rule induction. FMCG-Fast Moving Consumer Goods Fast Moving Consumer Goods (FMCG) sector Products which have a swift turnover and relatively low cost are known as Fast Moving Consumer Goods (FMCG). Fast Moving Consumer Goods (FMCG) sector includes the companies dealing with the all consumable items (other than groceries/pulses) that one needs to buy at regular intervals FMCG products are also known as Consumer Packaged Goods(CPG).These are items which are used daily, and so have a quick rate of consumption, and a high return. Examples of FMCG commonly include a wide range of repeatedly purchased consumer products such as toiletries, soaps, cosmetics, oral care products, shaving products and detergents, as well as other non-durables such as glassware, bulbs, batteries, paper products, and plastic goods. FMCG may also include pharmaceuticals, consumer electronics, packaged food products etc. Categories FMCG can be categorized in three segments as 1. Household items as soaps, detergents, household accessories, etc, 2. Personal care items as shampoos, toothpaste, shaving products, etc and finally 3. Food and Beverages as snacks, processed foods, tea, coffee, edible oils, soft drinks etc. Characteristics: Indian FMCG sectors significant characteristics can be listed as strong MNC presence, well established distribution network, intense competition between the organized and unorganized players and low operational cost. Easy availability of important raw materials, cheaper labor costs and presence across the entire value chain gives India a competitive advantage. The following are the main characteristics of FMCGs i. From the consumers' perspective: Frequent purchase Low involvement Low price ii. From the marketers' angle: High volumes Low contribution margins Extensive distribution networks High stock turnover Though the absolute profit made on FMCG products is relatively small, they generally sell in large quantities, so the cumulative profit on such products can be substantial. Growth of FMCG The mushrooming middle class Indian population, as well as the rural sector, presents a huge potential for this sector. The FMCG sector in India is at present, the fourth largest sector with a total market size in excess of USD 13 billion as of2012. This sector is expected to grow to a USD 33 billion industry by 2015 and to a whooping USD 100 billion by the year 2025.[1]In India the easy availability of raw materials as well as cheap labour makes it an ideal destination for this sector. There is also intense competition between the organized and unorganized segments and the fight to keep operational costs low. Indian Consumer Spending Pattern which affects FMCG sector growth Figure Figure 2: Indian Consumer Spending Behavior Figure 1: FMCG Segment Share ISSN No: 2250-3536 E-ICETT 2013 15
There are some factors which affects the growth of this sector. Following are some of them: Increasing rate of urbanization. Rise in disposable incomes, resulting in premium brands having faster growth and deeper penetration. Innovative and stronger channels of distribution to the rural segment, leading to deeper penetration into this segment. Increase in rural non-agricultural income and benefits from government welfare programmes. Investment in stock markets of FMCG companies, which are expected to grow constantly. Challenges: Following are some the challenges for this sector: Increasing rate of inflation, which is likely to lead to higher cost of raw materials. The standardization of packaging norms that is likely to be implemented by the Government by Jan 2013is expected to increase cost of beverages, cereals, edible oil, detergent, flour, salt, aerated drinks and mineral water. Steadily rising fuel costs, leading to increased distribution costs. The present slow-down in the economy may lower demand of FMCG products, particularly in the premium sector, leading to reduced volumes. The declining value of rupee against other currencies may reduce margins of many companies, as Marico, Godrej Consumer Products, Colgate, Dabur, etc who import raw materials. SWOT analysis of FMCG sector: Strength Low operational costs Presence of established distribution networks in both urban and rural areas Presence of well-known brands in FMCG sector Favorable governmental Policy Foreign Direct Investment(FDI) Weaknesses Lower scope of investing in technology Opportunities Untapped rural market, changing life style Rising income levels, i.e. increase in purchasing power of consumers Large domestic market with more population of median age 25 High consumer goods spending Threats Removal of import restrictions resulting in replacing of domestic brands Tax and regulatory structure Rural demand is cyclical in nature and also depends upon monsoon. TOP most companies i FMCG[4] Following are some of the top earning Companies in FMCG sector: Hindustan Unilever Ltd. ITC (Indian Tobacco Company) Nestlé India GCMMF (AMUL) Dabur India Ltd eckitt Benckiser Cadbury India Britannia Industries Ltd. Procter & Gamble Hygiene and Health Care Marico Industries Ltd. Colgate-Palmolive (India) Ltd. Gillette India Ltd. Godfrey Phillips Henkel Spic Johnson & Johnson Nirma Ltd Godrej Consumer Products Data Mining Technique There are different approaches to classify the data like we have parametric approaches, semi-parametric approaches, and nonparametric approaches. There are several techniques in Data Mining. We can apply any technique according to the need, Following are some of the Techniques of Data Mining. Association Rule Mining Apriori Algorithm FP Growth Classification Decision Tree Artificial Neural Networks Bayesian Classification Clustering Hierarchical Partitioned Grid ISSN No: 2250-3536 E-ICETT 2013 16
K means is the simplest and most classical clustering method that is easy to implement. The classical method can only be used if the data about all the object is located in the main memory, the method is called K means because each of the K clusters is represented by the mean of the object which is called the centroid within it. I t is also called centroid method. The k-means method uses the Euclidean distance measure, which appears to work well with compact clusters. If instead of the Euclidean distance, the Manhattan distance is used the method is called the K-means method. The k-means method may be described as follows: 1. Select the number of clusters. Let this number be k. 2. Pick k seeds as centroids of the k clusters. The seeds maybe picked randomly unless the user has some insight into the data. 3. Compute the Euclidean distance of each object in the dataset from each of the centroids. 4. Allocate each object to the cluster it is nearest to based on the distance computed in the previous step. 5. Compute the centroids of the clusters by computing the means of the attribute values of the objects in each cluster. 6. [Optional] One may decide to stop at this stage or to split a cluster or combine two clusters heuristically until a stopping criterion is met. Following is the flowchart for the K-means algorithm: data mining we have done preprocessing on data. Besides the stock information, we have collected Data about the audited information of each of these companies. This audited information includes Net Sales / Interest Earned/Operating Income, Other Income, Total Income, Expenditure, Interest, Profit Before Depreciation and Tax, Depreciation, Profit before Tax, Tax, NetProfit, Cash EPSetc. Figure 4 shows the original data table which is having ambiguities. Figure 4: Audited Reports of Company for last four years Besides the Audited data of the companies we have also collected the stock detail of following 10 FMCG sector companies for the last 10 years. 1. Colgate-Palmolive (India) Ltd. 2. Godrej Consumer Products Ltd. 3. Hindustan Unilever Ltd. 4. United Breweries Ltd. 5. Nestle India Ltd. 6. Vikas WSP Ltd. 7. Jyothi Consumer Products Ltd. 8. Agro Dutch Industries Ltd. 9. LT Foods Ltd. 10. Farmax India Ltd. Following figure 5 displays the details of Colgate Palmolive india ltd. Data Preprocessing For this research we collected the data from BSE website. As the data in its original format is cannot be used directly for ISSN No: 2250-3536 E-ICETT 2013 17
We have collected stock data for 10 FMCG companies. Datafields are High Value, Low Value, Closing Value of last 10years i.e from 2002 to 2012. This is displayed in followingfigure 6 Figure 7 Graphical Representation of Combined data of 10 years As we can see in figure 4 that EPS Earning Per Share isgiven for all stock. From the information of EPS and Closingprice we can formulate the P/E Ratio for each stock.p/e Ratio calculation is done by following formula. P/E Ratio = Closing Rate / Cash EPS. Earnings per Share is a total profit of the company devide bythe total share of the company.by applying this to all stock information we have derivedfollowing table of P/E Ratio of all stocks. This is calculatedand displayed in Figure 8. Following figure shows the graphical view of the stockinformation of 10 years records of 10 FMCG companies ISSN No: 2250-3536 E-ICETT 2013 18
Now from all row data of stock information and audited information, we have perform preprocessing and prepared training data set. After a preprocessing on the all data,we have formed the training table which is displayed in following Figure 9. Technical Analysis Now from the training data set of table 1.6 in first step wehave started to apply K means algorithm as described beforeto get the classes as Positive Growth, Zero Growth andnegative Growth. These classes may have differentcompanies from our training companies. Let us find out theresult of K means algorithm. As we can see in the Figure 10, data set retrieved afterimplementation of algorithm that there six companies which areunder Class C1, Only One Company Under class C2 and ThreeCompanies under class C3. Here C1,C2 and C3 are Three clusters as Positive Growth, ZeroGrowth and Negative Growth Companies respectively. This cluster information are displayed in following figure 11. Now we have derived clusters and have identified FMCG companies under each cluster. This information is required to identify the strong association between the growth rate and other required fields. These other required fields have to be identified by applying association rule on the detail dataset. Now let us consider different fields which may affects the growth rate of the Company. Following are some of the fields from the company information. Net Sales / Interest Earned / Operating Income Profit Before Depreciation and Tax Depreciation Net Profit Equity Capital Reserves Operating Profit Margin Net Profit Margin Cash EPS P/E Ratio ISSN No: 2250-3536 E-ICETT 2013 19
Out of all these financial fields we have identified that Cash as Net Profit, Reserve, Cash EPS of any company makes effect on the High Price, Low Price and Closing price of that year but due to competition in FMCG sector and lack of new innovation in products, companies Net Profit and Reserves are affected in a negative way. Due to the reduction in Net Profit, Reserves of the companies affected which simultaneously affect the EPS of the company. After analysis of consecutive four years audited reports, we came to the conclusion to consider Net Profit and Reserves besides the stock price of the company. So we have collected these two data for all above mentioned companies according to their growth rate. Following table shows the detail. As above figure shows the relative information of the Net Profit, Reserves and its corresponding growth rate. From the above mentioned table we can identify the maximum, minimum and average amount of Net Profit and Reserves and its corresponding Growth Rate of the Company. As we can HUL Hindustan Unilever is having highest rate ofnet Profit and Reserves and stood in cluster Positive Growth.Same way Agro Dutch has smallest Net Profit and Reserveand stood in Negative Growth.So we can conclude on following Two Points: 1. Net Profit, Reserves effect the growth rate. 2. by analyzing these two factors and current trends will behelpful to identify the growth rate from Zero to Positive andnegative to zero and vice versa is also true. accurateresult Expectation Maximization Method(EM Method ) canbe used. References 1) L.K. Soon and Sang Ho Lee, Explorative Data Mining onstock Data 2) Dattatray P. Gandhmal,Ranjeetsingh B. Parihar,Rajesh V. Argiddi An Optimized Approach to Analyze Stock market using Data Mining Technique, International Conference on EmergingTechnology Trends (ICETT) 2011 3) Jiawan Han, MichelineKamber Data Mining Conceptsand Techniques 4) Sukhvir Singh, Jagdeep Singh Association Rules and Mining Frequent Itemsets using Algorithms, International Journal of Computer Science &Engineering Technology (IJCSET) 2011 5) G.K.Gupta, Introduction to Data Mining with CaseStudies, Second Edition 6) Gebouw D, B-3590 Diepenbeek, Belgium Building an Association Rules Framework to Improve ProductAssortment Decisions 2004 7) http://info.shine.com/article/fmcg/fmcgindustryoverview/4261/cid780.aspx 8) http://en.wikipedia.org/wiki/fastmoving_consumer_goo ds 9) http://www.indiabulls.com/securities/market 10) http://entranceexam.net/forum/generaldiscussion/ list- top-50-fmcg-companies-india- 565386.html#ixzz28VM2A8wR Conclusion This paper gives the approximation of the growth rate andidentified the other criteria which may affect the growth rateof the companies. As FMCG sector having many fundamentaladvantages and having some problem too. For this, It isnecessary to identify the growth rate of the company and thendecide to invent in that. Here we have used Kmeans withassociation rule for rule induction but for better and ISSN No: 2250-3536 E-ICETT 2013 20