Free Report On Project Report

Introduction

In the daily work of the head of the sales department a great place is addressed to issues related to the planning of sales for the coming periods. It is necessary to consider the impact of various factors that may have an impact, both positive and negative. In this paper we analyze the list of such factors and give a brief description how they affect sales volume. Increase in sales compared to the growth of the market is the first criterion of sustainability of the business. If your company's sales growth is not satisfactory, we can confidently say that in your process of promotion and sale is a limiting factor. This may be the rules of procedure or sales figures to increase sales, a commercial offer, promotion methods, qualification or staff incentives, etc.

Problem Statement

The purpose of this assignment is to explore which factors are significantly associated with the sales volume in fashion stores. According to the conditions of the assignment, there was collected a data set of 400 observations. Each observation represents a Dutch fashion store indicators in 1990. There are five measures of the data is given - five quantitative variables, which are represented below in a table:
As it was mentioned above, we are investigating which factors are associated with sales volume. In this data we have “Sales” variable which is considered as dependent variable. The other four variables will be examined whether they affect “Sales” or not. However, the choice of these variables is not groundless. The explanation why these variables are included in our research is given below.
Sales staff is the link between the company and consumers. In the minds of most buyers, the sales manager is the "face" of the company. Therefore, it must be prepared by a qualified addressing each customer to be able to analyze them and always be ready to offer the right solution. To do this, he needs to know not only the analytical marketing techniques are particularly significant in the higher levels of trade management, but also be able to apply them. Thus, it is natural to assume that the number of employees may affect the sales volume.
Economic analysis of labor resources is especially important in market conditions. Since the survival of the company in a difficult competition depends on the ability to rapidly reconfigure production, in which the crucial role played by human resources company. The volume of production of goods depends not only on the number of employees as the number of labor expended on the production, a certain amount of working time. Therefore, the analysis of working time is one of the most important components of the analysis staff. In an analysis of the use of time to test the validity of production targets, studied the level and quality of their performance, identifies loss of working time and their causes, as well as identified reserves of increasing efficient use of time, developed the necessary arrangements. From the security of the enterprise human resources, efficiency of their use, as well as on the effective use of time depends on the efficiency of the company. Inefficient use of time leads to a decrease in production and sales, respectively, the company does not receive part of the profits. Decrease in loss of working time, without the additional costs lead to increased productivity and increased production. The number of hours worked is a good measure of productivity.
Sales per square meter should be monitored in order to understand how to effectively use trade area. It is known that the trade area, if it is managed effectively, boosts revenues of commercial enterprise to 20-30%. The effectiveness of retail space also shows sales per square meter commercial or network. According to this indicator can be compared to stores that sell very different products. Trade area to work effectively, it first must wisely choose to buy or rent, given the type of trade, which is expected to be engaged in this trade area, as well as its geographical location.

Analysis

One of the most common measures of association between the variables is Pearson’s correlation coefficient. Correlation reflects the fact that a change in one trait are in accordance with changes in some other features. Correlation indicates a causal relationship changes two features that, strictly speaking, is not always the case, as the correlation methods do not reveal this reason, and only indicate the presence of a match. Variables can be not only in mutual dependence on each other, but both depending on any third impact which is not included in the viewing area. The method of linear correlation (Pearson correlation) is used to determine compliance with the measures of the two variables, expressed quantitatively, in other words - for numerical values. This is a parametric method, which (as well as other parameter) requires compliance with the distribution of the attribute in the normal distribution.

Consider the following scatter plots and pair-wise correlation coefficients:

We are interested in the strength of association between Sales and other variables. According to the graphs, we can’t conclude that there is a really strong association between Sales and any of other variables, because the points on the graphs are not located near some trend line. The coefficients of correlation support this suggestion. At 5% level of significance, there are weak linear association between Sales and Nfull, HoursW and Ssize. The association vs. Nfull and HoursW is positive, the association vs. Ssize is negative. The association between Sales and Npart is not significant. It seems that Npart does not have a significant impact on Sales, that’s why we will not include this variable into regression equation.
However, the next step, according to conditions of the assignment is to construct linear regression based on Sales as dependent variable and HoursW and Ssize as independent variables.

Run the multiple regression analysis of Sales vs. HoursW and Ssize:

The obtained regression equation has the following form:
Y=5133.59+37.528X1-22.145X2

Where

Y-SalesX1- HoursWX2- Ssize
Results
The coefficient of correlation between Sales and Npart is only 0.05 with p-value of 0.318. This is evidence that there is no significant correlation between the variables. There is no linear association between number of part-time workers and sales per meter. This factor should not be put in a linear multiple regression equation, because it will not improve the model. The coefficient of this variable will be insignificant and the factor should be removed from the equation.
In my opinion, the regression equation should consist of Sales as dependent variable and HoursW, Ssize and Nfull as independent variables. It is possible, that there are some other factors which are significantly associated with Sales, but weren’t explored in this study.