Major Statistical Techniques Applied in Businesses - Research Paper Example

Major Statistical Techniques Applied in Businesses I. Introduction The term Statistics actually stands for a set of techniques which are intensively used for the purpose of collecting different sorts of data along with organizing them according to the purposes and finally analyze and interpret them to make some inference or draw conclusion on any issue. It is not that statistics deals with only quantitative data which can be expressed by numerical values, but also on many occasions statistical techniques are applied on qualitative data. In the areas of business and economics, statistical techniques are applied quite often for the purpose of analyzing and interpreting various issues related to business and economic decision making. In any business firm, statistical analytical tools provide great help to the management in their decision making. Statistics are really essential in the decision making procedure of the businesses as it enhances the understanding regarding the sources of variation in certain variables and uncovers different types of pattern and relationships among different variables from the business data. Statistical analyses of data are very important for strategy making for the business houses. In the same way to understand the pattern of relationship between certain variable of the economy, statistical techniques come to great help. (Downing and Clark, 1-2; Kazmier et al. 1) For the purpose of analysis of data, mainly two types of statistical techniques are applied in areas of businesses and economics, for instance, descriptive statistics and inferential statistics. Each of these categories includes various types of techniques. In case of descriptive statistics, those tools, which mainly summarize the available numerical data and describe them to make the interpretation easy, are used. On most of the occasion, the method of descriptive statistical analysis includes various types of graphical techniques and is mainly based on computational analysis. Inferential statistical analysis employed on those cases when business or economic decisions are required to be made under uncertainties. Inferential statistical analysis includes those techniques which help to make any decision regarding the entire population on the basis of an observed sample, not on the basis of the data for the entire population. In case inferential statistical, probability concepts are required to be incorporated in the techniques used for the purpose of decision making as the decision are made under uncertain conditions. When statistical characteristics of data are measured on the basis of a sample then, then the estimated values of those characteristics are known to be as sample statistics. On the other hand, when the same statistical characteristics are measured on the basis of the entire statistical population, then those estimated characteristics are known to be as population parameters. Census is the method which is applied to measure the characteristics of a defined population. But very often the values of population parameters are predicted on the basis of a sample. The technique of sampling is very often used to uncover or control the sources of variation in the data. (Downing and Clark, 1-2; Kazmier et al. 3; Daniel and Terrell, 2-3) In the areas of business and economics, the application of the statistical tools is of different types. For example, on many occasion the classical statistical methods are applied. These methods were developed in order to analyze sample data set and make inference about the population on the basis of the sample. In case of classical statistical methods no personal judgment on the selected data is incorporated during the time of analysis as well as at the time of making inferences about the population characteristics. These methods simply assume that the sample has been collected from a stable population. Another type of application is called decision analysis. Unlike in case of classical statistics, the decision analysis techniques take into account personal judgments. These methods put emphasis on incorporating judgments of the managers into the statistical analysis process. And the final category of application includes the techniques of statistical process control. These methods are applied when it is expected that the resulting output may not show a stable feature. (Kazmier et al. 2; Snider, 2) Very often managers need to examine the relationship between two variables, for example the relationship between certain managerial policy and employee efficiency, or they need to assess the impact of some strategy or certain market condition on their performance, or they need to forecast future performance in terms of profitability or sales on the basis of current and historical data, or to simply evaluate performance of the firm in long run or in short run on the basis of present and past data set. For all these purposes, the statistical techniques that are widely used are simple descriptive statistical tool, correlations, and regressions. It would be nice to briefly describe these statistical tools. (Downing and Clark, 4; Kazmier et al. 1) II. Major statistical techniques applied in Businesses A. Descriptive Data Analysis: Very often to build some ideas regarding a variable, the technique of data analysis is applied. Particularly when there arises a need of conducting a quantitative research, descriptive data analysis seems to be a very use full tool for gathering primary information regarding the variable under consideration. For example, if a manager is to evaluate the historical performance of the firm in terms of its revenue generation, then he has to first collect data on revenues for a particular period of time. For example he can take a sample of ten years. Now he can simply applies the technique of descriptive data analysis to infer about the past performances of the firm. For analyzing the data it is necessary to summarize them properly so that some ideas can be formed regarding the values of the variables in the data set and how these variables vary. Descriptive statistics mainly include the size of the sample collected for the purpose of study, the minimum and maximum values of the variables in the data set, average values of the variables as well as some measures of dispersion. (Downing and Clark, 9; (“Statistical Thinking for Managerial Decisions”) Among all these descriptive statistics, the two very crucial statistics that are mostly encountered by the researchers are the measures of averages which are mainly known as the measures of central tendency in the statistical language and measures of variation which are called measures of dispersion. Measures of central tendency are of different types like mean, median or mode each of which gives average values of the variables in the data set. But the procedure of calculating averages in each of these cases is different. Each of these measures has their own advantages and disadvantages. On the other hand measures of dispersion which measures the extent of variation in the values of the variables in the sample around their averages include Range, Standard Deviation, Quartile Deviation etc. (Downing and Clark, 9-10) B. Correlation analysis: On various occasions it becomes necessary for a strategy maker of a firm or for an economist to examine what kind of relationship holds between two variables in reality. It is not that what theory states about the relationship between two variables will always hold in practice. For example, very often employees’ participants in the entire decision making procedure of the firm is considered to be helpful for increasing efficiency of the workers. But there is no guarantee that it will definitely enhance the productivity of the employees. Whether the strategy of increasing employees’ participation within a firm will be considered as an effective strategy depends on the extent of the relationship between employee participation and employee productivity. If these two are related in a significantly positive way for one particular firm then the employee participation could be an effective efficiency enhancing strategy for that firm. But for another firm the relationship may be negative and for that firm this strategy should not be followed. Therefore for making a decision regarding certain strategy statistical technique of measuring correlations are widely applied in the business world. (Downing and Clark, 356-358) In the language of statistics, correlation can be defined as follows: Suppose X and Y stand for two random variables. Also suppose that the values that these two variables assumes in the selected sample come from random experiment in the sense for the given population the sample values for these two variables have been chosen in a random way where every item in the population had equal probability of getting selected for the sample. Now the covariance of these two variables can be represented as follows- Cov (X, Y) = E{[X - E(X)][Y - E(Y)]}; where E(X) and E(Y) stand for the expected or mean values of X and Y, respectively. Now assuming that the variances, i.e. squares of standard deviations, have positive values, then the correlation between these two variables can be defined as follows: r(X, Y) = Cov(X, Y) / [sd(X) . sd(Y)], where ‘sd’ stands for standard deviation. ‘r’ is defined as the correlation coefficient. If the value of r is positive on the basis of the given data then the two variables are said to be correlated positively in the sense that an increase in the value of X will be associated with the increase in the value of Y or vice versa. For example if X stand for the level of employee participation in corporate decision making and Y stands for the value of employee productivity then a positive value of r will indicate that increase in the level of employee participation will be associated with an increase in the level of employee productivity. Correlation can be measured by applying various procedures including Pearson Correlation measure, Spearman’s method of correlation and Point-Biserial correlations. Each of these is calculated by applying different techniques. (“Statistical Thinking for Managerial Decisions”; C. Regression analysis: Regression technique is one of the most important statistical techniques used by the decision makers for the purpose of strategy making. Regression is very important for the purpose of forecasting and for a business firm forecasting is very important as it reveals future potential of the firm and helps in current decision making. Apart from this regression technique is also very helpful in estimating impact of certain factors on some variables. For example, if a manager wants to find out how the markets demand of the product of its company is responsive to a particular economic condition, then regression analysis can be carried out. (Downing and Clark, 3341-343) The statistical technique of regression helps in predicting or estimating the value of one variable from the values of other variables. As far as the prediction for future is concerned, time series regression analysis can be conducted. A regression analysis generally starts with a hypothesis regarding how one variable is related another variable. Then the regression equation is estimated where the dependent variable is on the left hand side of the equation and the independent variable or variables are on the right hand side. The equation is estimated on the basis of the values of the variables given in the sample data set. Now if the coefficient of the independent variable in the equation appears to be positive and close to 1, then it can be said that the independent variable highly influences the dependent variable in a positive way or vice versa. (Downing and Clark, 341-343) Regression analysis can be of different types, such as simple linear regression which taken into account only one independent variable, multiple regression which considers more than one independent variables, etc. for estimating the regression equation mainly least square method are applied as it is the most convenient way of estimation. This paper only provides a very precise analysis of business statistics. Only some of the major statistical technique applied in business and economics have been discussed here precisely. No elaboration has been made regarding various procedures that are applied under each types of statistical analysis. So there lies further scope of research on different types of statistical techniques applied in business, there evolution over time, there uses for empirical testing and so on. III. A brief graphical explanation An important statistical technique applied to study some variable is the use of different graphs. If one tries to have a quick glance on the relationship between two variables then scatter diagram can be used. Suppose a manager tries to find out how the sales volume is related to the gross domestic product of the nation (GDP). The following table gives some hypothetical values of sales and GDP for five years. (Kazmier et al, 7) Year Sales(in thousand billion) GDP(in thousand billion) 2004 25 500000 2005 31 550000 2006 35 570000 2007 37.5 615000 2008 42 667000 The values of sales and GDP for each of the year can be plotted a s point on the scatter diagram as follows: The above graphical presentation shows that the higher values of sales volume (series 1) are associated with higher values of GDP. So the manager can infer that increase in GDP cause market demand for their product to rise. IV. Conclusion Application of various statistical techniques has become an integral part of the process of business decision making. Any strategic or managerial decision of a business organization is based on its past performances as well as on the expectations regarding the future. Managers need to set their strategy in such a way that available data can justify those decisions. Here comes the importance of the application of statistical techniques. Today every decision in any business organizations are made on the basis of the results coming from statistical analysis of available data. Works Cited 1. Arsham, Hossain. “Statistical Thinking for Managerial Decisions”. n.d. Retrieved from http://home.ubalt.edu/ntsbarsh/Business-stat/opre504.htm#rcorrIationCovar on 8th July, 2009. 2. Downing, Douglas., and Clark, Jeff. Business Statistics. Barron's Educational Series. 2003. 3. Daniel, Wayne W. and Terrell, James C. Business statistics: basic concepts and methodology. Houghton Mifflin, 1986 4. Kazmier, Leonard J., Fulks, Daniel L. and Staton, Michael K. Business statistics: based on Schaum's outline of theory and problems of business statistics. McGraw-Hill Professional. 2003. 5. Snider, Joseph L. Business statistics: a book of cases and materials. McGraw-Hill Book Company, inc. 1932 Read More