INTRODUCTION:

It’s not just the new cars who face a strong market in the UAE, but also the used cars have always been very much in demand. “The used car market in the UAE is worth around US$2.5 billion in business and it is growing by a strong 15 per cent annually,” according to Kerem Kuyucu, the managing director of Carmudi Middle East, an online marketplace for new and pre-owned cars. (Scott, 2014) Predicting the price of a used car is both an important and an interesting matter. According to data obtained from the Roads and Transport Authority, the number of light vehicles registered between 2009 and 2014 is more than a million which is a jump of about 80% of the cars registered solely in 2009. With respect to the economic conditions since 2009, the sale of second hand and reconditioned cars has ever since been on an increase. Predicting the resale value of a car is not a an easy task. It is common knowledge that the value of the used cars will depend on a number of factors. The most important being the age of the car, its make (and model), the origin of the car (i.e the original country of the manufacturer), its mileage (referring to the kilometres it has run) and of course its horse power. Additionally, due to the increase in fuel prices, the average fuel consumption of the car is also a crucial factor. Sadly though, most of the people do not know the exact figure for the fuel their car consumes per km driven. Moreover, factors like the type of fuel the car consumes, the interior style, the braking system, acceleration, the volume of of its cylinders (measured in cc), safety index, its size, number of doors, body colour, weight of the car, size of the tyres used for the car, type of wheels and wheel caps, consumer reviews, the fame associated with the manufacturer of the car, its physical state, it has cruise control or not, whether it has a sun roof, incorporates a manual or automatic transmission, whther it belonged to an individual or a company as well as other options such as how good is the air conditioner of the car, what type is the sound system, whether it is has a power steering, cosmic wheels, GPS navigator, all such factors may influence the price of the vehicle. In the UAE certain other special factors which buyers afix importance to is the nationality of the previous owners, whether the car had been involved in any serious accidents and whether it is a lady-driven car. Furthermore, the look and the feel of the vehicle definitely add a lot to the value of the car. Therefore, we can see that the price of the car depends on a variety of factors. Unfortunately though, all of the information regarding each factor is not usually available and the buyer has to make the final decision of purchasing the car at a particular price based on a limited factors only. In this research paper we shall also be focusing only on some of the factors mentioned above. Regression analysis is a method of estimating the relationship between several variables. Sample data is usually collected to investigate the relationship between a dependent variable and one or more independent variables. This method includes many techniques for creating the best fit model by the analysis of the variables. Multiple regression is merely an extension of simple regression. Usually, a model is simply called an equation. Model can be used to predict weather, the performance of the stock market, sales, profits, river levels, and so on. (Zainodin & Khuneswari, 2009). Multiple regression can also be used for predicting the prices of houses as well as the value of used cars in the market. This paper shall be focusing on the latter use of multiple regression.

PROBLEM STATEMENT:

PURPOSE OF THE STUDY:

This study is being done on behalf of a market research company on the second hand car market in the UAE. The focal point of this study is to determine the market value of a specific model and manufacturer by taking into consideration certain factors related to the vehicle. The study shall also be concentrating on creating a model for the company, which it can use to estimate the market value of the car using limited number of factors. The manufacturer under focus in this study is Toyota and the car is Land Cruiser, taking into account the models from 2009 till 2014.

DEFINITIONS OF TERMS:

LITERATURE REVIEW:

In our attempt to understand how the second hand car market in the UAE works, we must be able to breakdown how the car models and their prices are analyzed by their buyers. There are certain definable characteristics and attributes to cars that contribute to the overall appeal and market value that a given car elicits. Essentially, any type of good or commodity can be viewed as a package with many different characteristics that add or subtract to the overall value of that particular good. This is the same for car market as well. A car is simply a combination of characteristics (such as color, price, model, mileage, etc.) that all together contribute in some measurable way to the ultimate value that a particular buyer places on that car. To analyze the varying prices of the second hand cars in relation to their different characteristics, in this paper we shall use multiple regression analysis along with descriptive analysis and hypothesis testing.

MULTIPLE REGRESSION:

Multiple regression is a generalization of the simple linear regression analysis. Simple regression analysis could analyze a relationship between dependent variable with a single independent variable. The same idea was used to analyze relationship between a dependent and two or more independent variables. (Nikolopoulos, Goodwin, Patelis, & Assimakopoulos, 2007) suggested that multiple lregression is a common choice of method when a forecast is required and where data on several relevant independent variables are available. This technique has been widely used to produce forecasts in a wide range of areas and there is evidence that it is often used by companies to derive forecasts of demand from marketing varaibles and various macroeconomic measures. Multiple regression has been successfully used in many business applications. For example, (Evans & Olson, 2003) studied the NFL data, it would be logical to say that the number of Games WON would not only on Yards Gained but also on the other variables like Takeaways, Giveaways, Yards Allowed and Points scored. Multiple linear regression is a popular method for producing forecasts when data on relevant independent variables is available.

DESCRIPTIVE ANALYSIS:

Descriptive statistics are used to describe the basic features of the data in a study. They provide simple summaries about the sample and the measures. Together with simple graphics analysis, they form the basis of virtually every quantitative analysis of data. Descriptive Statistics are used to present quantitative descriptions in a manageable form. In a research study we may have lots of measures. Or we may measure a large number of people on any measure. Descriptive statistics help us to simplify large amounts of data in a sensible way. Each descriptive statistic reduces lots of data into a simpler summary. For instance, consider a simple number used to summarize how well a batter is performing in baseball, the batting average. This single number is simply the number of hits divided by the number of times at bat (reported to three significant digits). A batter who is hitting .333 is getting a hit one time in every three at bats. One batting .250 is hitting one time in four. The single number describes a large number of discrete events.

STATISTICAL HYPOTHESIS TESTING:

It is common knowledge that researchers usually wish to demonstrate that the phenomenon in question is present (i.e., reject the null hypothesis in the favor of the alternative one). However, there are instances in which researchers do have a priority, theoretically justified reasons to hypothesize formal, statistical null relationships. (Cohen, 1990; Cortina & Dunlap, 1997; Cortina & Folger, 1998; Greenwald, 1975, 1993). Support for positing and testing null relationships between variables of interest is offered by Greenwald (1993), who noted that “scientific advance is often most powerfully achieved by rejecting theories. A major strategy for doing this is to demonstrate that relationships predicted by a theory are not obtained, and this would often require acceptance of a null hypothesis” (p. 421). Others have argued that the tenability of the null hypothesis is as legitimate a goal of research as is demonstrating the tenability of any alternative hypothesis (Chow, 1996; Cortina & Folger, 1998; Frick, 1995; Nickerson, 2000). We do believe that theoretically based arguments that lead researchers to predict null relationships between their research variables of interest are justified. The formal statistical procedure for performing a hypothesis is to state two hypothesis and to use an appropriate statistical test to reject one of the hypotheses and therefore accept (or fail to reject) the other one. The first hypothesis is usually referred to as the Null hypothesis because it is the hypothesis of no effect or no difference between the populations of interest. It is usually given the symbol H0. On the other hand, the second hypothesis is usually the Alternative Hypothesis by statisticians, but since it is often the hypothesis that the researcher would like to be true, it is also sometimes referred to as the Study Hypothesis or Research Hypothesis. This is give the symbol H1 OR HA. The Alternative Hypothesis states that there is an effect or that there is a difference between the populations.

HYPOTHESIS ANALYSIS:

Furthermore, in this study we conducted hypothesis tests on some of the variables of our report. Paired sample t-tests were performed on the model year and the price of the respective car models. The following tables show the results of the tests performed in the SPSS Statistical Software.

Paired Samples Statistics

Mean

N

Std. Deviation

Std. Error Mean

Pair 1

Model Year

2011.70

80

1.672

.187

Selling Price

171152.50

80

45514.126

5088.634

The above table shows that for a mean of 2011 (for the model year), with the standard deviation at 1.672, the selling price is calculated to be at a mean of 171152.50 with a standard deviation of 45514.12.

Paired Samples Test

Paired Differences

t

df

Sig. (2-tailed)

Mean

Std. Deviation

Std. Error Mean

95% Confidence Interval of the Difference

Lower

Upper

Pair 1

Model Year – Selling Price

-169140.800

45512.738

5088.479

-179269.164

-159012.436

-33.240

79

.000

The above table shows that overall mean of the model year and the selling price is 169140.8 with a standard deviation of 45512.73. These tables were obtained by the SPSS statistical software by applying paired-samples t-test to compare the model year and the selling prices of the Land Cruisers discussed in this report. The results thereby, show that the mean selling price is 169140.8 for these cars with a standard deviation of 45512.73. The value of the t-test is 33.240 and the degree of freedom is 79. These results suggest that the model year really does affect the price of the Land Cruiser. Specifically, our results suggest the older the car’s model year the lower the price of the car will be and vice versa. A Pearson product-moment correlation coefficient was computed to assess the relationship between the model year of the Land Cruisers and their respective selling prices.

Correlations

Model Year

Selling Price

Model Year

Pearson Correlation

1

.830**

Sig. (2-tailed)

.000

N

80

80

Selling Price

Pearson Correlation

.830**

1

Sig. (2-tailed)

.000

N

80

80

**. Correlation is significant at the 0.01 level (2-tailed).

The above table suggests that there was a positive correlation between the two variables, r = 0.830, n = 80, p = .000. A scatterplot summarizes the results (Figure 1). Overall, there was a strong, positive correlation between the model year of the cars and the selling prices. Increase in model year (latest model) results in a higher selling price of the car.

Bibliography

Chow, S. L., 1996. Statistical significance: Rationale, validity and utility. Thousand Oaks, California: Sage Publications. Cohen, J., 1977. Statistical power analysis for the behavioral sciences. Rev ed. New York: Academic Press. Cohen, J., 1990. Things I have learned (so far). American Psychologist, Volume 45, pp. 1304-1312. Cortina, J. M. & Dunlap, W. P., 1997. On the logic and purpose of significance testing. Psychological Methods, Volume 2, pp. 161-172. Cortina, J. M. & Folger, R. G., 1998. When is it acceptable to accept a null hypothesis: No way, Jose?. Organizational Research Methods, Volume 1, pp. 334-350. Evans, J. R. & Olson, D. L., 2003. Statistics, Data Analysis and Decision Modeling. 2 ed. New Jersey: Prentice Hall. Frick, R. W., 1995. Accepting the null hypothesis. Memory & Cognition, Volume 23, pp. 132-138. Greenwald, A. G., 1975. Consequences of prejudice against the null hypothesis. Psychological Bulletin, Volume 82, pp. 1-20. Greenwald, A. G., 1993. Consequences of prejudice against the null hypothesis. In: K. G & L. C, eds. A handbook for data analysis in the behavioral sciences: Methodological Issues. Hillsdale, New Jersey: Lawrence Erlbaum, pp. 419-448. Nickerson, R. S., 2000. Null hypothesis significance testing: A review of an old and continuing controversy. Psychological Methods, Volume 5, pp. 241-301. Nikolopoulos, K., Goodwin, P., Patelis, A. & Assimakopoulos, V., 2007. Forecasting with cue information: A comparison of multiple regression with alternative forecasting approaches. European Journal of Operational Research, 180(1), pp. 354-368.