Gojan Natalia EMREI-115

Laboratory work nr. 3

Time series and forecasting
1.) Compute the time series indicators for your data (GDP for France and Moldavian GDP) and make a description of the evolution of the GDP.
Table 1 GDP Moldova
GDP (Gross Domestic Product) 32,03 37,65 44,75 53,42 62,92 60,42 71,88 Absolute modification current 5,62 7,1 8,67 basic 0 5,62 12,72 21,39 1,175 1,189 1,194 Growth coefficient current basic 1 1,175 1,397 1,668 117,55 118,86 119,37 Rate of growth, % current basic 100 117,55 139,71 166,78 17,546 18,858 19,374 Modification rate, % current basic 0 17,546 39,713 66,781

Year 2004 2005 2006 2007

2008 2009 2010

9,5 -2,5 11,46

30,89 28,39 39,85

1,178 0,96 1,19

1,964 117,78 196,44 1,886 96,027 188,64 2,244 118,97 224,41

17,784 96,441 -3,973 88,636 18,967 124,41

Table 2 GDP France

GDP (Gross Domestic Product) Absolute modification current 99,405 131,51 121,77 basic 0 99,405 230,91 352,68 1,056 1,071 1,061 Growth coefficient current basic 1 1,056 1,131 1,2 105,64 107,07 106,11 Rate of growth, % current basic 100 105,64 113,11 120,02 5,6438 7,0676 6,1122 Modification rate, % current basic 0 5,6438 13,11 20,024

Year 2004 2005 2006 2007

2008 2009 2010

1 761,3 1 860,7 1 992,2 2 114,0 2 191,2 2 175,1

2 214,4

77,263 -16,17 39,4

429,94 413,77 453,1

1,037 0,993 1,018

1,244 103,65 124,41 1,235 99,262 123,49 1,257 101,81 125,73

3,6548 24,411 -0,738 23,492 1,8097 25,727

I created Table 1 and Table 2 and computed the Time series indicators (absolute modification, coefficient growth, rate of growth and modification rate) in Excel using data from http://www.oecd.org and http://www.statistica.md/. The absolute modification is the difference between the GDP of the comparable year and the GDP of the previous year.

The growth coefficient (current) is the division of the GDP of comparable year by the GDP of previous year. The growth coefficient (basic) is the division of the GDP of comparable year by the GDP of the basic year. The growth rate is the growth coefficient (current and basic) multiplied by 100%.

In order to make a description of the evolution of the GDP in Moldova and France, I created two line charts that show the evolution of GDP in both countries. Chart 1
GDP 80 70 60 50 40 30 20 10 0 2004 2005 2006 2007 2008 2009 2010 Year 37.65 32.03 44.75 53.42 62.92 60.42 GDP Trendline

Evolution of GDP in Moldova (2004-2010)

71.88

On Chart 1 we may distinguish 5 phases of the Business Cycle in Moldova: 1. A very long Recovery or Expansion (2004-2008). From 2004 to 2007 the actual output coincides with the trend level. A recovery is characterized by an increase in national income and output. Unemployment falls but consumption, investment and imports begin to increase. 2. Peak (2008) with GDP of 62, 92 bln. USD. There is a positive output gap because the actual output is above its trend level. National income is high. 3. Recession (2008-2009), influenced by the global financial crisis of 2007. Output and income fall leading to a fall in consumption and investment. 4. Trough (2009) - economic activity is low comparable to surrounding years. 5. Recovery (2009- 2010) output begins to increase very fast to 71, 88 bln.USD in 2010.

Chart 2
GDP 2 500.0 2 114.0 2 000.0 1 500.0 1 000.0 500.0 0.0 2004 2005 2006 2007 2008 2009 2010 Year 1 860.7 1 992.2 1 761.3 GDP Trend line

Evolution of GDP in France (2004-2010)

2 191.2 2 175.1 2 214.4

On this chart we can also distinguish 5 phases of the Business Cycle in France: 1. Expansion or Recovery (2004-2008). From 2006 to 2008 the level of output is above the trend line. There are expansions and rise in economic activities. When demand starts rising, production increases and this causes an increase in investment. There is a steady rise in output, income, employment, prices and profits. 2. Peak or Boom (2008) of 2191, 2 bln. USD 3. A short Recession (2008-2009). During a recession period, the economic activities slow down. When demand starts falling, the overproduction and future investment plans are also given up. There is a steady decline in the output, income, employment, prices and profits. 4. Trough (2009). 5. Low Expansion or Recovery (2009-2010).

Conclusion: The time series of both countries indicate cyclical fluctuations. If we compare the evolution of the GDP in Moldova and France we can see that from 2004 to 2010 both countries had 5 phases of the business cycle in the same years. The peak/boom was in 2008 and the trough in 2009 for both of countries, influenced by the global financial crisis.

2.) Construct the time linear regression. Consider the X values as numbers from 1, 2 , and the Y values - the GDP values for 2004-2010 years. Use the INTERCEPT and SLOPE functions to determine the parameters of the regression. Or use the special tool in Excel. Construct the linear regression equation. I have chosen to use the INTERCEPT and SLOPE functions to determine the parameters of the regression. Time linear regression (Moldova):
Table 3

Year 2004 2005 2006 2007 2008 2009 2010

GDP, bil.$ (Y) 32,03 37,65 44,75 53,42 62,92 60,42 71,88

Time variable 1 2 3 4 5 6 7
Slope= Intercept= 6,545 25,687

Y=sx+i s- slope i-intercept Y=6,545x+25,687 - linear regression equation

Time linear regression (France):

Table 4

Year 2004 2005 2006 2007 2008 2009 2010

GDP, bil.$ (Y) 1 761,3 1 860,7 1 992,2 2 114,0 2 191,2 2 175,1

2 214,4

Time variable 1 2 3 4 5 6 7

Slope= Intercept=

78,112 1731,7

Y=sx+i s-slope i-intercept Y=78,112x+1731,7 - linear regression equation

3) Calculate the forecasting for GDP (France) in the next year (2011 year). Forecasting of the GDP of France using the linear regression equation:
Table 5

Year 2004 2005 2006 2007 2008 2009 2010 2011

GDP, bil.$ (Y) 1 761,3 1 860,7 1 992,2 2 114,0 2 191,2 2 175,1

2 214,4

Time variable 1 2 3 4 5 6 7 8
Y (2011) =78,112x+1731,7=78,112*8+1731,7=2356,578

2356,58

4.) Calculate the forecasting for GDP (Moldova) the in next year (2011 year). Forecasting of the GDP of Moldova using the linear regression equation:
Table 6

Year 2004 2005 2006 2007 2008 2009 2010 2011

GDP, bil.$ (Y) 32,03 37,65 44,75 53,42 62,92 60,42 71,88 78,047

Time variable 1 2 3 4 5 6 7 8
Y (2011) =6,545x+25,687 = 6,545 * 8 + 25,687=78,047

5) Draw the line plot for GDP Select in plot the line GDP and then use the menu Chart/ Add trend line. Construct the linear equation, power equation, exponential equation. What plot (graph) showing the tendency is better?

Moldova: Chart 3 Evolution of GDP in Moldova (2004-2011)- using linear equation

Y 90 80 70 60 50 40 30 20 10 0 1

y = 6.545x + 25.687 R = 0.9755

71.88 62.92 53.42 37.65 44.75 32.03 60.42

78.047

GDP, bil.$ (Y) Trendline

Time variable 2 3 4 5 6 7 8

Chart 4 Evolution of GDP in Moldova (2004-2011)- using exponential equation

100 90 80 70 60 50 40 30 20 10 0

y = 25.687e0.1534x R = 0.8989

78.047 53.42 44.75 32.03 37.65 62.92 60.42 71.88 GDP, bil.$ (Y) Trendline

Chart 5 Evolution of GDP in Moldova (2004-2011)- using power equation

71.88 62.92 53.42 60.42 37.65 44.75 32.03 Time variable 1 2 3 4 5 6 7 8 GDP, bil.$ (Y) Trendline 78.047

Y 90 80 70 60 50 40 30 20 10 0

y = 29.652x0.4369 R = 0.9604

France: Chart 6 Evolution of GDP in France (2004-2011) -using linear equation

Y 2 500.0 2 114.0 2 000.0 1 500.0 1 000.0 500.0 0.0 0 2 4 6 8 10 1 860.7 1 992.2 1 761.3 GDP, bil.$ (Y) Trendline 2 175.1 2 191.2 2 214.4 2356.578 y = 78.109x + 1731.7 R = 0.9338

Time variable

Chart 7 Evolution of GDP in France (2004-2011) using exponential equation y = 1731.7e0.0398x

R = 0.9209 2 114.0 2 000.0 1 500.0 1 000.0 500.0 0.0 1 2 3 4 5 6 7 8 Time variable 1 860.7 1 992.2 1 761.3 GDP, bil.$ (Y) Trendline 2 191.2 2 175.1 2 214.4 2356.578

Y 2 500.0

Chart 8 Evolution of GDP in France (2004-2011) using power equation

Y 2 500.0 2 114.0 2 000.0 1 500.0 1 000.0 500.0 0.0 1 2 3 4 5 6 7 8 1 860.7 1 992.2 1 761.3 GDP, bil.$ (Y) Trendline 2 191.2 2 175.1 2 214.4 y = 1732.9x0.1358 R = 0.9623 2356.578

Time variable

A trend line indicates the general pattern or direction of a time series data. Trend lines are used to graphically display trends in data and to analyze problems of prediction. Such analysis is also called regression analysis. By using regression analysis, you can extend a trend line in a chart beyond the actual data to predict future values. When you want to add a trend line to a chart in Microsoft Excel, you can choose any of the six different trend/regression types. The type of data that you have determines the type of trend line that you should use.

A trend line is most reliable when its R-squared value is at or near 1. When you fit a trend line to your data, Excel automatically calculates its R-squared value. Analyzing the charts for GDP of Moldova we can see that r is more close to 1 if we use the linear equation (Chart 3). On the other hand, for France, the best chart is Chart 8 using power equation. So the graph that shows the tendency better is Chart 3 (using linear equation) for Moldova and Chart 8 (using power equation) for France.

6) Compare the evolution of the two countries (Moldova and France). Use the average absolute modification and average modification rate. Average absolute modification (Moldova):

Average modification growth rate (Moldova):

Average absolute modification (France):

Average modification growth rate (France):

7) Conclusions about the evolution and possibilities of the time series analysis. In exercise 5 we can see that all graphs are showing an increasing tendency in the evolution of the gross Domestic Product of Moldova and France. After computing the forecasting for 2011 using the linear regression equation, average absolute modification and the coefficient k, we can compare them in the following table:
Forecasting (2011) using: Linear regression k 78,047 78,45 81,58 2356,578 2299,44 2302,97

State: GDP of Moldova, bln.USD GDP of France, bln. USD

The GDP of Moldova is higher if we use the k coefficient to forecast for the 2011 year, and lower with the linear regression method. The GDP of France is highest using the linear regression method and lower using the average absolute modification. In both cases there is an increasing tendency in the evolution of GDP from 2010 to 2011.

10