# Cases on Simple Regression

CASE 1 : Fiscal Deficit – Oil Imports Subsidies on oil tend to put an immense pressure on the Fiscal Deficit of our country. This case therefore assumes that as Oil Imports increase, the Fiscal Deficit should increase and vice versa and therefore checks for correlation between the two variables for the last two calendar years.
Date Jun-12 May-12 Apr-12 Mar-12 Feb-12 Jan-12 Dec-11 Nov-11 Oct-11 Sep-11 Aug-11 Jul-11 Jun-11 May-11 Apr-11 Fiscal Deficit in Billion Rupees 488.73 743.91 671.96 161.6 586.38 539.21 276.43 463.6 261.99 72.87 447.7 661 319.27 560.65 746.61 Oil Imports in Billion Rupees 710.95 816.41 695.33 800.6 645.29 763.48 615.12 632.47 553.38 513.47 566.24 574.9 595.52 590.23 576.99

On performing regression on the above data, the critical values show the following output.

As can be seen. the ‘Rsquare’ value of 0. For example. . an increase in the fiscal deficit can be due to increase in subsidies to fertilizers for agriculture.06119 is too less to show any correlation. It can also be because of subsidies to other sectors influencing the fiscal deficit. Conclusion : No Correlation The reason for this can be due to reasons such as delay or early payment of subsidies such that it gets reflected in different months.

83 79.1 80.48 79.79 79.48 110. the critical values show the following output.42 79. Date 16-Oct-12 15-Oct-12 14-Oct-12 12-Oct-12 11-Oct-12 10-Oct-12 9-Oct-12 8-Oct-12 7-Oct-12 5-Oct-12 4-Oct-12 3-Oct-12 2-Oct-12 1-Oct-12 Dollar Index 79.03 79.08 79.18 116.9 Oil Price 115 115 115.48 110.48 110.6 112.42 79.43 80.17 110.86 79.17 116. The data for the month of October is used for this case.11 114.68 79.77 79.17 115.86 80.32 112.48 On performing regression on the above data. Any fluctuations in the value of dollar as indicated by its index could influence the price of dollar.CASE 2 : Oil Spot Price – Index of Dollar Value(DXY) The price of the oil is backed to the dollar.17 112. This case therefore tries to find any correlation between the two variables. .

3 158. CASE 3 : Wholesale Price Index(WPI) – Consumer Price Index(CPI) Both the WPI and the CPI are indicators for Inflation in India with differences in them in terms of how they are calculated.06072 is too less to show any correlation.8 164. One of them could be the overwhelming effect of demand and supply.4 152.As can be seen.2 154.9 154.5 161 159. WPI uses the wholesale prices while CPI uses consumer prices.2 163. . the critical values show the following output.3 157. However. Conclusion : No Correlation There could be several reasons for the lack of correlation. Date 12-Jul 12-Jun 12-May 12-Apr 12-Mar 12-Feb 12-Jan 11-Dec 11-Nov 11-Oct 11-Sep 11-Aug 11-Jul 11-Jun 11-May 11-Apr 11-Mar 11-Feb 11-Jan WPI 164. the ‘Rsquare’ value of 0. are they correlated? This case tries to find out.5 148. The forces of demand and supply could influence the Oil prices a lot more than the dollar value such that the variables show no correlation.1 152.9 163.1 149.2 153.4 157 156.7 157. The differences in their measurements give different outputs.1 148 CPI 656 646 638 633 625 621 618 618 621 619 615 610 604 598 592 587 585 584 589 On performing regression on the above data.

936828 shows a strong correlation.As can be seen. . The residuals also lie within the range of +2 to -2 indicating no stray values.0. the ‘Rsquare’ value of 0.

both CPI and WPI are correlated. Their equation is: Y = 5. This also means that both of them measure Inflation appropriately.246X .Conclusion : Correlation Exists In spite of the differences in their measurements.621 + 0.

5 5.16667 FDI 1472 1476 4105. However there are several factors which play a role in the GDP moving up or down.166667 5.266667 12.9 7.3 9.333 1281.667 On performing regression on the above data.333 4480.9 9. The data is for the last 3 calender years.Cases on Multiple Regression CASE 1 : GDP – IIP – FDI GDP is one of the most important variables for a country to showcase its growth.45 0. This case tries to find a relation between variables such as IIP and FDI and how it affects GDP.333 1506 1732. Date Q2 2013 Q1 2013 Q4 2012 Q3 2012 Q2 2012 Q1 2012 Q4 2011 Q3 2011 Q2 2011 Q1 2011 GDP 5.566667 1.966667 7.433333 7.333 1684 1898. .533333 6. the critical values show the following output.966667 7.1 6.8 8.7 7.4 IIP 0.3 6.3 8.533333 2.667 1935.

However.0. the ‘Rsquare’ value of 0.936828 shows a strong correlation.As can be seen. Both IIP and FDI are necessary for a nation to show increased and improved growth. . the P-value for FDI isn’t optimal and hence should be avoided. The residuals also lie within the range of +2 to -2 indicating no stray values Conclusion : Correlation Exists The above analysis indicates that all the 3 variables are correlated.

9 163.4 191.5 178.9 178.8 176.5 157.2 206.3 167.1 149.1 152.1 190.4 Fuel and Power 175. However.3 158.8 186. etc.7 192.3 197.2 209.3 On performing regression on the above data. fuel.6 161.3 186.5 161 159.CASE 2 : WPI – Food Inflation – Fuel and Power Inflation WPI when calculated involves a lot of internal variables. The figure of WPI includes the inflation for food and non.5 199.2 154. Date Jul-12 Jun-12 May-12 Apr-12 Mar-12 Feb-12 Jan-12 Dec-11 Nov-11 Oct-11 Sep-11 Aug-11 Jul-11 Jun-11 May-11 Apr-11 Mar-11 Feb-11 Jan-11 WPI 164. .9 154.5 148.2 163.2 197.1 207.8 188.2 193.4 157 156.6 160.9 196.2 153.3 192.6 153.8 179 181. the critical values show the following output.4 152.7 171.1 192. This case tries to correlate WPI with the food and fuel inflation.4 159.1 165.2 178.8 164. each of these does not have to be in line with the WPI figures.food articles.8 177.6 170 168. manufacturing goods.3 157.7 157.1 148 Food 212.7 177 172.5 151.

the ‘Rsquare’ value of 0.0. .936828 shows a strong correlation. The residuals also lie within the range of +2 to -2.As can be seen.

41975 + 0. . food and fuel inflation move in line with WPI and therefore can be considered as the influencing variables on WPI.2234 X1 + 0.4016 X2. A strong correlation exists between the variables and the equation is Y = 41.Conclusion : Strong Correlation As can be seen from the analysis.