GDP and Freight Forecast - Coursework Example

EXECUTIVE SUMMARY There is a linear relationship between freight forecast and several factors. The number of households leads to an increase in population leading to an increased demand for goods and services which leads to an increase in freight. The average of people employed also leads to more income hence increasing the GDP in the country leading to more demand and this directly affects freight as more goods and services are brought into the country. The above factors affect the purchasing power of the people and in turn affecting demand and supply leading to more or less goods being imported. All these data can be extrapolated and used to determine or forecast the amount of freight for the future by using the linear equation and using time series. This data is important for planners. Figure 2.0 Occurance 1 The graph above figure 1.0 shows that the estimated forecast on road freight estimated keeps increasing with time. This can be attributed to the increase in other factors e.g. The increase in GDP which is directly related with the increase in number of employed people in Victoria. However there is a slight drop in the estimated VIC-NSW road freight between 1982 and 1984, this is attributed to the drop in average number of people employed in Victoria figure 2.0 . A decrease in the number or employed people reduces the amount of money in the economy hence the reduction in amounts of goods coming into Victoria .This will reduce the number of vehicles supplying goods to Victoria due to the drop in demand. Consequentially the reduction of vehicles supplying goods will cause a drop in the estimate road freight in VIC-NSW road. Occurance 2 There were slight fluctuations in the transport fuel index between 1998 and 1999 and between 2008 and 2009. This is irregular and can attributed to external factors e.g. cost of fuel. An increase in the cost of fuel could lead to a drop in the fuel index as forecasting bodies would believe that most people wouldn’t buy fuel due to a sharp increase in the prices of fuel. A sharp increase in prices of fuel can be caused by external economic factors e.g. global recession or instability in oil regions. GDP and freight forecast Increase in the GDP lead to the increase in the freight forecast. The increase is caused by an increase in the amount of money circulating in the economy. This increases the demand for goods due to increased disposable income. The increase in demand would make supplier transport goods to Victoria to counter the demand hence the increase in freight forecast. Freight forecast and transport fuel index There is a gradual increase in the transport fuel index. The transport fuel index is used to forecast the demand for fuel in the transport industry of a particular country. The increase in transport fuel index would mean an increase in the demand for fuel which in turn means and increase in the number of vehicles coming into a country. The transport fuel index is directly proportional to the freight forecast in Victoria as evidenced by the graph above. Freight forecast and estimated weekly earnings Increase in the estimated weekly earnings is directly proportional to the freight forecast .An increase in weekly earnings means an increase in income for the citizens of Victoria. This will lead to an increase in disposable income leading to an increase in demand for goods .Consequentially the increase in demand will trigger suppliers to transport goods and services into Victoria. Freight forecast and number of employed people. The freight forecast is directly proportional to the average number of employed people. An increase in the number of employed people increase the GDP of the country. An Increase in GDP increase the amount of money in the economy and the rate of business is also increased. This increase in demand and supply of goods and services will lead to an increase in the amount of vehicles coming in and going out which in turn increases the freight forecast of Victoria. Freight forecast and retail turnover The retail turnover is directly proportional to the freight forecast. Retail Turnover quantifies the purchases at the consumers’ place of expenditure. An increase in retails turnover is caused by an increase in the purchasing power of citizens in a country. The increase in purchasing power will lead to increased demand for goods and services. This will lead to increased transportation to provide for the demand. Freight forecast and number of households. The number of households is directly proportional to freight forecast. Increase in households means an increase in population. This means an increase in demand which will make suppliers transport goods to cover for the demand. The increased population will also increase the workforce hence goods will be exported from the country. Hence the increase in the forecast freight. Task 4 Variable R2 SEE t statistics F values X1 0.983372452 590.9882026 42.13354 1.44E-27 X2 0.93387024 1178.591165 20.60712 7.29E-19 X3 0.980457 640.7013 38.80863 1.5E-26 X4 0.973756 742.4651 33.3787 1.08E-24 X5 0.994145489 350.6792059 71.38104747 3.83E-34 X6 0.982391 608.1815386 40.92267 3.31E-27 Task 5 In all the variables the R2 is between 0.93 and 0.99 which shows a perfect fit of between 93% and 99%. This means that all these factors directly affect the forecast in freight. Since adjusted R2 is closest to one percent. The SEE Standard estimate error is used to show the accuracy of our predictions, as from the above X2 is the least accurate variable to use for our predictions. It will provide the least accurate predictions since it has a high standard deviation of 1178. This shows that transport fuel index doesn’t clearly predict the freight. Rationality test. It is done by using either the critical method or the p value approach. The critical approach uses the T stat value in the Anova table. While the P Value approach uses the P values. The P states that If the P-value is less than (or equal to) α, then the null hypothesis is rejected in favor of the alternative hypothesis. And, if the P-value is greater than α, then the null hypothesis is not rejected. In this case all our P values are less than 0.05 hence our results are statistically significant or rational. Task 6 Variable combination R2 SEE T stat P. VALUES 1 X1,x2 0.98839 499.4117 -0.12122 12.87504 3.412021 0.90435 1.61E-13 0.001919 2 X2,x3 0.977406 696.684 1.17E-05 0.519044 2.72E-09 1.17E-05 0.519044 2.72E-09 3 X3,x4 0.983095 602.6263 -4.74347 4.015521 3.213812 5.17E-05 0.000384 0.003203 4 x4,x5 0.994097 356.1187 0.207614 0.061942 10.01082 0.836982 0.951034 6.44E-11 5 X5,x6 0.995933 295.5925 -3.30099 9.924257 3.619073 0.00256 7.84E-11 0.001113 6 X1,x2,x3 0.982572 506.6012 0.17653 5.181218 2.223744 -0.42747 0.861148 1.69E-05 0.034408 0.672309 7 X2x3x4 0.982572 611.8838 -4.36434 -0.35934 3.171865 3.097602 0.000157 0.722036 0.003656 0.004406 8 X3x4x5 0.994416 346.3639 0.395699 1.629874 -0.43819 7.73219 0.695326 0.114328 0.664608 2.01E-08 9 X4x5x6 0.995844 298.7992 -2.64082 0.617162 5.607723 3.632287 0.013373 0.542115 5.29E-06 0.001116 10 X1,x2,x3,x4,x5 0.996729 265.0843 -4.08384 -2.92837 2.488596 4.923094 4.561032 0.000354 0.006844 0.019289 3.74E-05 9.91E-05 In all the combinations, the R2 values are closest to one hence all the combination will be closest to the line of fit or are assumed to predict accurately and correctly the freight forecast. The combination of x2,x3,x4 is the least accurate and predictive while the combination of x1x2,x3,x4,x5 is the most predictive using the R2 Values. The SEE Standard Deviation Errors range from 265 to 696.This show how far the equation deviates from the line of best fit. The higher the deviation the least accurate the predictions. In the combinations above most accurate is x1, x2, x3, x4, x5 and the least accurate is X2, X3, and X4 To test the rationality we use the P values approach and the critical approach. Using the p value approach most of the combinations have a p value of more than 0.05 hence they fail the rationality test. Some combinations e.g. (X4, X5) and (X1, X2, X3, X4, X5) P values of less than 0.05 Task 7 The most statistically and predictive equation should pass all the statistical tests. It should have a low standard deviation and a R2 which is closest to one. It should also have a p value of less than 0.05.Hence the most statistically and predictive combination is (X1,X2,X3,X4,X5). It has a P value of less than 0.05 in all variables and has the lowest standard deviation closest of 265 and the highest R2 that is closest to one of 0.99. Hence the most predictive equation Y=mx+c y=-9.24x3+3.31x4+0.14x5+11x6 -17211.7 Task 8 Since the relationship is linear between all variables, the excel function FORECAST to determine the other independent variables. The independent variables are as follows.   Estimated an forecast Vic-NSW road freight a(i) Gross Domestic Product at current dollars AU b Transport fuel index Melbourne c Estimated total weekly earnings Persons Victoria d Average of Total employed persons Victoria e Retail turnover Victoria f Estimated Total number of h'holds in Victoria g Year million tonne kms $ billion   $.c 000S $million 000s   Y X1 X2 X3 X4 X5 X6 1982 3818 184.05 33.7 276.38 1711 11165.8 1392.7 1983 3496 199.53 37 300.10 1681 12565.3 1406.4 1984 4183 223.99 39.5 343.00 1732 13436.6 1420.3 1985 4302 248.38 42.4 354.80 1780 14629.0 1434.3 1986 4739 271.49 46 377.05 1855 16424.0 1448.4 1987 4797 306.11 51.3 397.45 1917 17842.8 1466.9 1988 5300 347.21 53.2 418.98 1953 19128.8 1485.7 1989 5732 389.19 55.9 448.98 2047 20564.2 1503.5 1990 6002 413.46 60.9 478.08 2082 20967.0 1518.6 1991 6068 416.69 62.5 497.35 1978 21227.1 1572.2 1992 6136 432.06 64.4 504.08 1948 22035.7 1600.7 1993 6582 454.72 66.3 521.52 1927 23106.2 1628.1 1994 6957 482.46 68 548.15 1974 23593.4 1657.7 1995 7331 512 70 561.50 2043 25689.1 1674.3 1996 7999 542.91 71.4 573.65 2075 26700.7 1691.0 1997 8555 573.53 72.1 583.95 2086 28798.7 1712.7 1998 9123 604.86 70.4 595.40 2130 30117.0 1719.0 1999 9794 637.79 71.4 603.25 2150 33558.5 1735.0 2000 10451 686.24 76.9 618.25 2218 34282.1 1771.0 2001 10777 727.49 79.1 642.40 2258 37586.1 1817.0 2002 11350 778.75 80.2 687.90 2291 40823.7 1851.0 2003 12001 829.72 81.6 736.45 2338 43794.0 1884.0 2004 12601 892.99 83.6 744.15 2380 46929.3 1916.0 2005 13150 961.05 87 774.80 2451 47539.6 1946.0 2006 13749 1038.3 91.6 793.05 2509 50339.7 1929.0 2007 14545 1131.17 92.7 820.20 2594 53173.1 1965.0 2008 15668 1236.74 97.7 844.20 2652 55742.9 2002.0 2009 15958 1260.9 93.7 886.45 2678 59261.6 2038.0 2010 16328 1357.55 95.5 956.40 2758 61746.3 2076.0 2011 16882 1452.41 98.7 982.55 2814 63427.1 2154.4 2012 17715 1502.2 101.4 1011.75 2832 64220.9 2195.3 2013 18538 1554.6 103.2 1055.10 2857 65567.1 2237.0 2014  16953.7 1640.25 104.58 1091.278 2907 68076.67 2281.299 2015  17638.37 1729.96 105.93 1128.354 2957 70645.43 2327.086 2016  18347.99 1823.92 107.24 1166.348 3009 73274.75 2374.413 2017  19075.21 1922.34 108.52 1205.285 3061 75966.07 2423.331 2018  19827.21 2025.42 109.76 1245.187 3115 78720.86 2473.894 Task 9 y=-9.24x3+3.31x4+0.14x5+11x6 -17211.7 y=-9.24*1091.10+3.31*2907+0.14*68076.67+11*2281.30-17211.7 y=16953.7 Y=-9.24*1128.35+3.31*2957+0.14*70645.43+11*2327.09-17211.7 y=17638.37 Y=-9.24*1166.35+3.31*3009+0.14*73274.75+11*2374.41-17211.7 y= 18347.99 Y=-9.24*1205.29+3.31*3061+0.14*75966.07+11*2423.33-17211.7 y=19075.21 Y=-9.24*1245.19+3.31*3115+0.14*78720.86+11*2473.90-17211.7 y=19827.21 Task 10 Since the data is linear I will use the linear trend line in trend line analysis. The data changes progressively with time with a noticeable pattern with a minimum deviation. Hence linear trend analysis is the best method. Task 11 Using forecast Excel Using trend analysis  16953.7 17744.6  17638.37 18231.8  18347.99 18719  19075.21 19206.2  19827.21 19206.2 The results obtained from the two methods are quite different but the trend analysis give a clear picture since logically we expect an increase in freight due increase in the factors described. The excel forecast gives an unexplainable drop at the beginning while trend analysis continues with the trend. I will therefore recommend the trend analysis to planners. Task 12 The cost of fuel can be used to determine the freight between NSW and Victoria. An increase in cost of fuel will reduce the freight. This can be attributed to suppliers trying to cut cost by reducing trips or using larger vehicles. Another factor will be alternative means of transport. The efficiency of other transport models e.g. rail system will reduce the freight on the road as bulky goods will be transported by rail or water. REFERENCES Bell, Kathy. Regression. Owen Sound, Ont.: Northern Sanctum Press, 2009. Print. MLA formatting by BibMe.org. Read More