• Join over 1.2 million students every month
• Accelerate your learning by 29%
• Unlimited access from just £6.99 per month

# Estimate a consumption function for the UK economy explaining the economic theory and statistical techniques you have used.

Extracts from this document...

Introduction

Estimate a consumption function for the UK economy explaining the economic theory and statistical techniques you have used.

Consumption has been considered as the most important single element in aggregate demand, accounting for almost 66% of GDP in 1989. Therefore, it is essential that the level of consumption be predicted accurately, for even a small percentage error may lead to a large absolute error. Another reason why consumption such important is that the marginal propensity of consume is one of the factor which is used to determine the size of the multiplier. This will influence the changes of investment and government spending. Moreover, as saving ratio, another factor that used to predict the behavior of consumers, has fluctuated rapidly during these years, so the consumption become more important. Thus, many economists attempt to develop many theories and equations to predict consumers’ expenditure. Therefore, in this project, there are two main theories and their equations will be given. Some data from Economic Trends Annual Supplement and graphs will be used to estimate the performance of the model on the basis of some econometric tests.

Keynes, John Maynard introduced consumption function in 1936. In The General Theory of Employment Interest and Money, Keynes pointed out that

“ We should therefore define what shall call the propensity to consume as the functional relationship, between Y, a given level of income and C the expenditure on consumption out of that level of income…The amount that the community sends on consumption depends (i) partly on the amount of its income, (ii) partly on other objective attendant circumstances, and (iii) pa        rtly on the subjective needs and the psychological

Middle

ct=c0+c1yt+c2ct-1+c3πt

where πis the inflation rate.  Use the same data as the previous equation to gain the equatipon.we can gain new equation as below:

LC=+1.001*LY_1-0.002921*p

According to the tests which display in the Table 6, we can see that the coefficients of both parameters are significant as the value of t-test are quit high. But DW just 0.14, below than 2, which means not good. Figure 6(a) and (b) shows the same conclusion as DW.

Thus, another factor which called error correction should be added to the consumption. As in the short run consumption will not equal the desired proportion of income. Thus, assume that consumption towards their consumption to their long-term goal. Then, consumption function in this case is:

Δct=c0+c1Δyt+c2st-1+c3πt

Use the same ways to gain the test result to estimate the consumption function. The following consumption function can be generate from the data is:

DLC=+0.01119+0.6508*DLY+0.1092*LS_1-0.001078*p

In Table 6, the result of the tests shows that all the absolute value of the t-test are bigger than 2, which are good. The highest value in this test is DLY, which is 7.23. It is means that the coefficient of DLY is significant in this equation. The same conclusion can be gained from the other tests. Such as the p-test, the higher the value of t-test the lower the value of p-test. Thus, the consumption function which incorporating inflation and error correction is successful. The predictions from this equation are shown in Figure 7(a). The equation performs very well from 1950 to 1982. This can be shown in the Figure 7(b) as well. In the same period, the graph is random.

Conclusion

60;LC =  + 0.6873 + 0.9453*LYP

(SE)     (0.128)  (0.0101)

Figure3 ### Table 4

EQ (4) Modelling LC by OLS (using project12min.xls)

The estimation sample is: 1950 to 2002

Coefficient  Std.Error  t-value    t-prob  Part.R^2

LC_1                0.632198     0.1034     6.11       0.000   0.4278

Constant            0.206148     0.1075     1.92       0.061   0.0685

LY                  0.351899     0.09701    3.63       0.001   0.2083

R^2                 0.998055     F(2,50) =    1.283e+004 [0.000]**

log-likelihood      138.489    DW                      0.858

no. of observations 53         no. of parameters           3

mean(LC)              12.6755    var(LC)              0.161805

LC =  + 0.6322*LC_1 + 0.2061 + 0.3519*LY

(SE)     (0.103)       (0.108)  (0.097)

### Figure 4 ### Table 5

EQ( 5) Modelling LC by OLS (using project12min.xls)

The estimation sample is: 1950 to 2002

Coefficient    Std.Error    t-value    t-prob    Part.R^2

LY_1               0.999415      0.0004450    2246.       0.000    1.0000

log-likelihood       94.4484    DW                       0.46

no. of observations        53   no. of parameters           1

mean(LC)              12.6755    var(LC)              0.161805

LC =  + 0.9994*LY_1

(SE)     (0.000445)

Figure 5 Table 6

EQ( 6) Modelling LC by OLS (using project12min.xls)

The estimation sample is: 1950 to 2002

Coefficient     Std.Error    t-value   t-prob  Part.R^2

LY_1                1.00082          0.0006735     1486.      0.000   1.0000

P                   -0.00292117      0.001097     -2.66       0.010   0.1221

log-likelihood        97.9004       DW                       0.41

no. of observations        53       no. of parameters           2

mean(LC)              12.6755        var(LC)              0.161805

LC =  + 1.001*LY_1 - 0.002921*P

(SE)     (0.000674)   (0.0011)

Figure 6 ### Table 7

EQ( 7) Modelling DLC by OLS (using project12min.xls)

The estimation sample is: 1950 to 2002

Coefficient  Std.Error  t-value  t-prob Part.R^2

Constant            0.0111950   0.004826     2.32   0.025   0.0990

DLY                  0.650824    0.09000     7.23   0.000   0.5163

LS_1                 0.109243    0.05125     2.13   0.038   0.0849

P                 -0.00107845  0.0003864    -2.79   0.007   0.1372

R^2                  0.632806  F(3,49) =     28.15 [0.000]**

log-likelihood        158.397  DW                       1.27

no. of observations        53  no. of parameters           4

mean(DLC)           0.0265864  var(DLC)          0.000404345

DLC =  + 0.01119 + 0.6508*DLY + 0.1092*LS_1 - 0.001078*P

(SE)     (0.00483)  (0.09)       (0.0512)      (0.000386)

### Figure 7 Table 8

EQ(8) Modelling DLC by OLS (using project12min.xls)

The estimation sample is: 1950 to 2002

Coefficient  Std.Error  t-value  t-prob Part.R^2

Constant            0.0152043   0.004323     3.52   0.001   0.2049

DLY                  0.429792    0.09556     4.50   0.000   0.2965

LS_1                0.0650428    0.04600     1.41   0.164   0.0400

P                 -0.00178033  0.0003788    -4.70   0.000   0.3152

HP                0.000917289  0.0002263     4.05   0.000   0.2550

R^2                  0.726458  F(4,48) =     31.87 [0.000]**

log-likelihood        166.199  DW                       1.55

no. of observations        53  no. of parameters           5

mean(DLC)           0.0265864  var(DLC)          0.000404345

DLC =  + 0.0152 + 0.4298*DLY + 0.06504*LS_1 - 0.00178*P + 0.0009173*HP

(SE)     (0.00432)  (0.0956)     (0.046)        (0.000379)  (0.000226)

### Figure 8 Table 9

EQ(9) Modelling DLC by OLS (using project12min.xls)

The estimation sample is: 1952 to 2002

Coefficient  Std.Error  t-value  t-prob Part.R^2

Constant            0.0176739   0.005014     3.52     0.001   0.2164

DLY                  0.393545    0.1146        3.43    0.001   0.2075

LS_1                0.0426819    0.05511     0.774    0.443   0.0132

P                 -0.00166234  0.0005008     -3.32    0.002   0.1967

HP                 0.00100082  0.0002372      4.22    0.000   0.2835

U                   -0.101787     0.1912    -0.532    0.597   0.0063

R^2                  0.717132  F(5,45) =     22.82 [0.000]**

log-likelihood        160.144  DW                       1.57

no. of observations        51  no. of parameters           6

mean(DLC)           0.0273824  var(DLC)          0.000387711

DLC =  + 0.01767 + 0.3935*DLY + 0.04268*LS_1 - 0.001662*P + 0.001001*HP- 0.1018*U

(SE) (0.00501)  (0.115) (0.0551) (0.000501) (0.000237)  (0.191)

### Figure 9  REFERENCE

#### All handouts from EQ1162

http://www.gmu.edu/departments/writingcenter/handouts/abstract.html

http://www.usfca.edu/economics/veitch/ECON621/Assignments/621Assign1%20F01.doc

http://www.iupui.edu/~econ/courses/whcin.pdf



This student written piece of work is one of many that can be found in our AS and A Level Core & Pure Mathematics section.

## Found what you're looking for?

• Start learning 29% faster today
• 150,000+ documents available
• Just £6.99 a month ## Here's what a teacher thought of this essay

3 star(s)

The author recognises the importance of consumption to the UK economy and models it well. However, explanation of statistics is poor, even though the conclusions are generally accurate. The application of theory and comparison of models based on theory is a good approach for this essay.
The conclusion doesn't seem to discuss which models were good and which were bad, which was the point of the essay. It does however talk about which other factors could be considered.
3 stars.

Marked by teacher Nick Simmons 28/02/2012

Not the one? Search for your essay title...
• Join over 1.2 million students every month
• Accelerate your learning by 29%
• Unlimited access from just £6.99 per month

# Related AS and A Level Core & Pure Mathematics essays

1.  ## C3 Coursework - different methods of solving equations.

5 star(s)

Because this only works out the value to 3 decimal places, I will use the Newton Raphson Formula to work it out to 5 decimal places. I will summarise it in a table like previously So I have to work out F' of my equation.

2.  5 star(s)

Moving Q closer to P will help me obtain the actual gradient of the curve. This applies to any value of x. In this next exercise; I will keep moving these two points together to obtain a gradient.

1. ## Determine an appropriate parabolic model that fits the data I collected of whirlybird wing ...

I decided to remove these from my final data. The result is shown below. Length (cm) Time (s) 14 1.29 12 1.7 10 2.27 8 2.33 6 2.1 4 1.65 2 0.99 Function: Y= a(X-b)2+c I will be referring to the letters in this function to explain changes.

2. ## Sequences and series investigation

If we look at the symetrical sides of the pattern and add up the number of squares we achieve a square number. Attempting to Obtain a Formula Through the Use of the Difference Method I will now apply Jean Holderness' difference method to try and find a formula. Pos.in seq.

1. ## Numerical solutions of equations

There has been no change in x-values between x7 and x8 for this number of decimal places. It is clear that another root has been successfully found using this method. My positive root is x = 0.724491959 (9 decimal places)

2. ## C3 COURSEWORK - comparing methods of solving functions

Therefore, we might miss the other two after finding the first one. This reason is that the two points are too close to each other. There are 3 root in the graph, 2 in the same interval [-3, -2] and 1 in the interval [0, 1].

1. ## Fractals. In order to create a fractal, you will need to be acquainted ...

In order to start the iteration, you need to start with a starting value, or seed, that takes the form of z0. Thus, iterating this starting value with the function yields: z1 = (z0)2 + z0 z2 = (z1)2 + z0 z3 = (z2)2 + z0 z4 = (z3)2 + z0 z5 = (z4)2 + z0 and so on.

2. ## Evaluating Three Methods of Solving Equations.

If the formula was a complicated one, this would be the easiest method to find its roots as long as the roots don't lie too close to each other, or would not show up easily. Newton- Raphson method This method requires you to find the derivative of the equation, which • Over 160,000 pieces
of student written work
• Annotated by
experienced teachers
• Ideas and feedback to 