Research Method for Business
A Statistical Test on PCE and GDP

Introduction

The objective of this paper is to present statistical test on two important variable – personal consumption expenditure (denoted as “PCE” in this essay) and gross domestic products (denoted as “GDP” in this essay). From the social science perspective, there are researchers argued that when the economy expand (i.e., when people earn more money), people are more likely to consume and spent more. Under such assertion, when the income of the public increased, they will spend more. The objective of this paper is to examine such assertion.

 

Two of the main variables used in this study are as follow:

  • PCE (i.e., represent the amount people spent in a particular year)
  • GDP (i.e., the total economy size, which can represent the total income earned of the public in a particular year)

 

The data will be obtained from Gujarati (2003). The full set of data is presented in Appendix.

Statistical Calculation

In order to perform calculation on the secondary data obtained from Gujarati (2003), statistical software, namely SPSS will be used. Firstly, the data of PCE as well as GDP for the United States are input into the software. In order to understand the nature of the data set, the minimum, maximum, mean and standard deviation of both the variables are calculated (presented in Table 1). Besides, the GDP data and the PCE data are plotted against year in the x-axis to understand the trend of GDP and PCE of US from year 1982-1996. However, these do not solve our problem, as the objective is to understand the relationships between GDP and PCE. If statistical significant relationship can be found, then we can conclude that GDP is indeed related to PCE. As such, the scatter plot of PCE against GDP is plotted. The “Regression” function in SPSS is used to generate a statistical model that related GDP to PCE.

Presentation of Data

The data generated from SPSS will be presented in this section. The interpretation and analysis of the output from the statistical software will be performed in the next section.

table-1

2

 

3

 

Analysis of Data

Table 1 shows the descriptive statistics of GDP and PCE. However, as both these variables represent the condition of people in United States from time to time, the mean value is of little value to us. In our study, our objective is to understand if when people earn more, they will spend more. Anyway, scatter plot in Figure 1 and 2 shows that both GDP and PCE of people from United States are increasing from the year 1982-1996. Apparently, there are linear relationships between GDP and PCE. Thus, in Figure 3, PCE is plotted against GDP. It is surprising that a roughly relationship between GDP and PCE exist. It is found that PCE is positively related to GDP. That means, when people earn more money, they will spend more. It seems like when people earn more, a portion/ percentage of their income will be spent. In order to generate model from the data set, regression analysis is performed. The results from regression are shown from Table 2 to Table 4.

Improvement

According to Gujarati (2003), the data is obtained from US governmental agencies. That means the data is pretty reliable. Thus, is can be reasonable believed that the secondary data used in the analysis is accurate. For this case, the use of primary data, by a single researcher is almost impossible. Both the variables, GDP and PCE that are collected by government, had gone through a complicated a long process. A single researcher simply does not have the financial resources, network, time and capability to perform data collection for GDP or PCE. The best a single researcher can do is to perform sampling. But sampling will be less accurate if compared to rigorous data collection process used by government. Thus, unless a team of researchers are supported by strong financial budget, it is not viable for them to collect primary data.

However, there are many ways in which the data used in the analysis above may not sufficient to form strong conclusion on spending habits of people. Firstly, the data collected is purely from US. We cannot be sure if people from other country may spend proportionally as well when their income increases. Do people from the East exhibit such tendency as well? This is an important question to be answer. Secondly, the data is quite old. Today is already 2011, and thus a more recent data of PCE and GDP up to the year of 2011 should be employed. Is that true that people still spend proportionally when their income increases today?

Conclusion

Overall, the analysis presented above is sufficient to conclude that when people earn more, they will spend more. The spending pattern is proportional to the level of income they earn. Such a statement is proven valid and accurate in the United States, from the year 1982-1996. However, we cannot conclude that such a scenario is applicable today, nor it is applicable in other countries. More study may be required to understand how people spend today. In fact, in the future study, the spending habits of people should be separated into different categories, such as spending on food, apparels, cars, luxury, education, entertainment and etc. Such a research will be more valuable as we can then understand what specific items people will spend on when their income increases.

References & Bibliography

Gujarati, D. N. (2003). Basic Econometrics (4th Edition). New York: McGraw Hill/ Irwin.

 

Appendix

Year Personal Consumption Expenditure GDP
1982 3081.5 4620.3
1983 3240.6 4803.7
1984 3407.6 5140.1
1985 3566.5 5323.5
1986 3708.7 5487.7
1987 3822.3 5649.5
1988 3972.7 5865.2
1989 4064.6 6062.0
1990 4132.2 6136.3
1991 4105.8 6079.4
1992 4219.8 6244.4
1993 4343.6 6389.6
1994 4486.0 6610.7
1995 4595.3 6742.1
1996 4714.1 6928.4

 

 

 

(Visited 47 times, 1 visits today)

About the author

Related Post

Leave a comment

Your email address will not be published. Required fields are marked *