Regression Analysis on Tata Tea
Regression Analysis on Tata Tea
1. History of Tea Industry
The discovery of tea in India was a serendipitous one. It was in the verdant valleys of Assam that the first tea in India was discovered during 1820's. The first English tea garden was setup by the British in 1837 at Chabua in Upper Assam. Then in 1840, the Assam Tea Company started the first commercial production of tea. Beginning in 1850's, the tea industry made large strides and Assam became a major tea producing region of the world. Since then tea has been cultivated successfully in different parts of India and has been a flag bearer for India at the global level. The tea industry has earned great kudos as a foreign exchange earner and due to its contribution to India's GNP. In all aspects of tea production, consumption and export, India has emerged as a world leader and has kept its leadership intact more or less for the last 150 years.
2. Trends of Tea Industry
Tea has been the pride of India. But recently it is been facing a lot of problems, some which are its own creation while others that are external. A decreasing demand for tea both in the international and Indian markets has been a cause of headache for the tea planters and others alike. The contribution of this industry to the general well being of the economy cannot be overstated. However this trend has been more pronounced in India than in the other competing countries because in around 2000 when the tea prices fell in India, there was a comparative increase in prices in the Indonesian markets. This indicates a vicious trend of distortions brought in by manipulations in auctions. Stipulations that bulk of the tea has to be sold at auctions, designed to promote fair competition, is resulting in the formation and favouring of cartels and thus prevent the normal forces of the market from operating. In these trying times Tata Tea has been a pioneer. Its “Jaago Re” has been able to induce social awakening in the minds of the youth. Apart from that with a slew of mergers, it has positioned itself as a global brand. On the other hand, Warren Tea is not a market leader with respect to volumes of tea produced but is regarded right at top with respect to quality.
3. Methodology
The methodology followed by me during the execution of this project involved of the following steps.
1. Studying the trends of the tea industry and trying to identify the different factors that might be affecting the industry.
2. Collection of data about Tata Tea and Warren Tea from different sources like indiastat.com, capitaline.com and others.
3. With help of Eviews software running regression analysis on the different factors and selecting or discarding them on the basis of the results.
4. Obtained the Regression Equation.
5. Interpretation of the different results like t statistic, F statistic etc obtained with the help of software.
6. Conclusion about the trends followed by the tea industry and future prospects discussion.
4. Experimentation
We did our experiments on data from 1995 2008.
Dependent Variable:
QUANTITY:
This variable is the quantity of tea demanded from Tata Tea.
Independent Variables:
PRICE:
This variable indicates the average price of the Tata Tea.
INCOME:
This variable represents the per capita income of India.
SUBS:
This variable is an indicator of the average price of tea in Warren Tea.
4.1 Results as obtained in Eviews
Dependent Variable:
QUANTITY
Method:
Least Squares
Date:
08/15/09 Time: 22:30
Sample:
1995 2008
Included observations:
14
Variable Coefficient Std. Error t Statistic Prob.
C 37214158 8854574. 4.202817 0.0018
PRICE 230801.6 177583.7 1.299678 0.2229
INCOME 2480.942 578.4157 4.289202 0.0016
SUBS 1063.962 733.8246 1.449886 0.1777
R squared 0.851088 Mean dependent var 63115929
Adjusted R squared 0.806415 S.D. dependent var 14752785
S.E. of regression 6490978. Akaike info criterion 34.44468
Sum squared resid 4.21E+14 Schwarz criterion 34.62727
Log likelihood 237.1128 F statistic 19.05129
Durbin Watson stat 2.052223 Prob(F statistic) 0.000185
Estimation Command:
=====================
LS QUANTITY C PRICE INCOME SUBS
Estimation Equation:
=====================
QUANTITY = C(1) + C(2)*PRICE + C(3)*INCOME + C(4)*SUBS
Substituted Coefficients:
=====================
SALES = 37214158 230801.6*PRICE + 2480.942*INCOME 1063.962*SUBS
This is the regression equation.
Year Quantity Price of Tata Tea(Rs.) Income(Rs.) Price of Warren Tea(Rs.)
1995 52,302,000.00 52.28 10149 51.487
1996 51,077,000.00 63.12 11564 59.62
1997 51,175,000.00 71.4 12707 65.56
1998 56,332,000.00 86.29 14396 87.54
1999 54,992,000.00 108.76 15625 85.75
2000 53,560,000.00 102.53 16555 94.12
2001 49,327,000.00 96.01 17823 93.48
2002 43,341,000.00 98.69 19040 76.67
2003 67,940,000.00 104.63 20989 78.83
2004 74,781,000.00 100.13 23241 64.59
2005 79,277,000.00 110.81 25788 86.91
2006 80,922,000.00 119.19 27784 75.51
2007 82,653,000.00 127.33 29382
87.43
2008 85,944,000.00 131.74 33283 91.71
4.2 Graphs and Interpretation of the results obtained
4.2.1 Graphs: (With the help of MSExcel)
The quantity of tea produced by Tata Tea produced does not follow a linear trend. There have been fluctuations in the way. As can be seen in the graph the production decreased almost to a new low in 2002 03. However after that it has been an upward swing. This spells good luck for the tea industry.
The price of Tata Tea has also seen a volatile trend. With the changing market demands its price has also been fluctuating. However all in all it has seen an increasing trend.
The per capita income of the Indian people has been rising ever since the economic liberalization of India took place during 1991 92. This trend is well documented in this graph. This increasing income has helped the Indian consumer to become a little more choosy while making their purchase decisions whereas it also gave them a chance to indulge themselves. This is the reason for an unsatiable demand in India despite different market dynamics that are applied.
The price of Warren Tea has been more volatile than Tata Tea and perhaps it might just be following the trend of its own constraints and inconsistencies rather than being a very strong competitor for Tata Tea.
5. Interpretation of results obtained from Eviews
5.1 Coefficient of determination(R squared):
The value of R2 which is 0.851088 indicates that approximately 85.11 percent of the variation in Tata Tea sales can be accounted for by the variation in the average prices of Tata Tea, per capita income of the population and the average prices of Warren Tea. This indicates the factors that have been taken by us can predict the trend followed by the sales of Tata Tea to a maximum efficiency of 85.11 percent.
5.2 Adjusted R squared:
In order to take into consideration the fact that the number of degrees of freedom declines as additional independent or explanatory variables are included in the regression, we need to calculate the adjusted R2. In our study, we found that the value of adjusted R2 is 0.806415. In our case the number of observations or sample data points (n) that we took is 14 and the number of coefficients (k) is 4. Hence the degrees of freedom is (n k) or (14 4) = 10. Thus the decrease in the value of adjusted R2 is understood. The proportion of the total variation in sales explained by the regression is approximately 80.64 percent.
5.3 Coefficients of the Regression Equation:
a. Coefficient of PRICE = 230801.6
This indicates that an increase in prices of Tata Tea by Rs.10 would result in a decrease in demand by 2308016kgs. It is an indication that the sales of Tata Tea are very sensitive to price fluctuations.
b. Coefficient of INCOME = 2480.942
This result shows that if there is an increase of Rs.1000 in per capita income of the people of India there would be a corresponding increase in sales by 2480942. Tata Tea is priced a little higher than most other brands. So perhaps people tend to view it as a superior good as compared to others like Warren Tea.
c. Coefficient of SUBS = 1063.962
This indicates that when there is an increase in the prices of Warren Tea by Rs.10 there is a decrease in the quantity demanded of Tata Tea by 10639kgs. It may be due to the fact at a higher price the consumer perceives Warren Tea as a better product and so it eats into the share of Tata Tea.
5.4 F statisticDegree of freedom of the numerator = 4 - 1 = 3
Degree of freedom of denominator= 14 - 4 = 10
So now we have to look at the F statistic table for 5 percent significance. Corresponding to 3df in numerator and 10df in denominator we find that the value of F statistic is 3.71. Thus the obtained value of F statistic which is 19.05129 is greater than the critical value. And so we reject at the 5 percent level of significance the null hypothesis that there is no statistically significant relationship between the independent variables (PRICE, INCOME and SUBS) and the dependent variable (QUANTITY).
5.5 t Statistic
The t statistic or t ratio allows us to know how significant any independent variable is on the dependent variable. So higher the value of t statistic, more confident we are that we can reject the null hypothesis that there is no significant relationship between the variables.
The value of t statistic as found in the t statistic table for 10df is 2.228.
a. PRICE and SUBS:
The t statistic value of both of these variables is less than the critical value. In case of PRICE it is understandable because there is a very large market for tea in India. People would consume tea even at the most adverse conditions and if it is a trusted brand like that of the Tatas then it is an added advantage. For SUBS it can be said that there are other companies both in tea industry as well as coffee industry who could be more suitable competitors to Tata Tea than Warren Tea.
b. INCOME:
The t statistic value for INCOME is 4.289202 which exceed the critical value. This indicates that per capita income is a significant factor that influences the sales of Tata Tea.
5.6 Standard Error of Regression:
The Standard Error of Regression shows how good a fit is the regression line to the observation or sample of points. In this case it is 6490978.
5.7 Durbin Watson Statistic:
This statistic gives us the value of correlation. However in this case it is not applicable because for looking up for the value in the DW table we require 15 or more observations.
6. Conclusion:
Studying the trends of the tea industry with respect to Tata Tea and Warren Tea I found out that what actually affects the sales of tea in case of Tata Tea was the increasing income of the Indian people. This gives us the indication that it is considered as a superior good by the consumer. The recent announcement by Tata Tea that it is going in for a strategic reorganization of its beverage business to focus on becoming a global market leader in the fast growing $160 billion ‘good for you' beverage market, indicates the well being of the company. The increasing sales volume is a clear indication of it.The factor of population which has acted in a big way in the regression analyses of different sectors does not seem to affect the sales of Tata Tea. This was discovered by me during experimentation when I found the t statistic value to be very low. Thus I dropped population as a variable for calculating demand. The Indian tea industry has been going through a lot of stress recently with decreasing sales volume and revenue, but now with the easing out of recession things can only get better.
References:
The sites followed by me apart from www.google.com while making this assignment are:
1) www.indiastat.com
2) www.capitaline.com
3) www.financialexpress.com
4) www.tatatea.com
5) www.wikipedia.com
