Quantitative Analysis
A parameter is a variable that is used to estimate the true population characteristics. In statistics, a parameter variable is derived from the sample data and used in describing the population characteristics. The population is composed of an infinite number of variables that are difficult to examine and make a sufficient conclusion. For instance, the total number of robust costumers may be a large number which can be difficult to interrogate and obtain the required information. Sample parameter helps in tackling these difficulties by examining a given number of customers and relating the observed data with the total population of customers. This makes easier to arrive to a conclusion when making decisions about population. Use of the parameter is cost effective, and time saving since collecting data from the total population may take a considerable amount of time. A well structured procedure is also needed when obtaining information from the total population. Test of parameter biasness and sufficiency should be done in order to arrive to the true population characteristics. A margin of error known as variance is provided since there are minimal variations that may arise whereby, these errors expected value is taken to be zero. The analyst should be able to evaluate on variables such as customers' tastes and preferences, close substitutes of the product with the market pricing level, costumers reaction on after sales services, as well as their par capita income. These variables among other influence the buying behavior of customers.
When an analyst is able to analyze these variables, an effective strategy can be made in order to meet the customers needs. A model can be made with each variable being allocated a specific coefficient in order to be used to estimate the total population characteristics. Parameters used should be unbiased, that is, their expected value should be equal to the population variables. Also, they should be efficient and sufficient in estimating the total customers' responses. This means that the variance from the expected value should be minimal as possible. When all these elements of parameters are met, then a quality decision is arrived at, as well as a clear way of making a decision. This will be able to explain the customers' behavior towards the marketing procedures, as well as other variables affecting their buying decisions.
An inference is the process of deriving logical conclusions from premises assumed to be true. Drawing of inferences by the business acts as a comparison of an expected outcome of the study from the actual conclusion made. They are taken to be the hypothesis to be evaluated by the study. The use of the parameters in studying the true population characteristics can be tested using a given significance level, testing on the acceptance or rejection of the hypothesis made. When business draws inferences in an attempt to assess its performance, as well as its customer relation strategies, inferences help business to establish the variance, which is the difference between the actual and expected variables, thus deciding whether the variance is adverse or favorable. A favorable variance indicates that the business is in the right trend whereas adverse shows the opposite of favorable variance. A business should rely on the inference when the expected outcome is almost equal to the actual outcome. This is a situation where the variance is favorable, or it is zero. When the test statistics are done, and the calculated statistics fall within the acceptance region, then the business should ascertain the correctness of the inferences made.
Test of difference is used to test the significance of the variance between the expected variable and actual results. In the process of estimating the population parameters, an estimation error occurs which may be a sampling error or the error of the content. These errors occur either due to failure of testing some variables in the sample or mathematical errors in making inferences. This makes the actual information obtained to have a variance from the expected value. A significance test needs to be done to establish whether these differences statistically influence the conclusions. An acceptance region is defined, and test of difference obtained is evaluated whether is in the region of acceptance or rejection. The acceptance of the error is when it does not statistically affect the true inferences made. On the other hand, a hypothesis is a proposed explanation of a situation. Hypothesis testing is done when an inference is made, and the correctness of the inferences needs to be ascertained. Two hypotheses are made that is, the null hypothesis and alternative hypothesis. Null hypothesis are always accepted unless enough evidence is provided to reject them. The alternative hypothesis tries to counter the null hypothesis. These hypotheses are tested at a given significance level when the standard error or deviation is given.
Both tests of difference and hypothesis testing are done to test the sufficiency, accuracy, as well as completeness of the inferences made by a business in conducting its research. Every analyst is required to do these tests in order to meet the required set objectives.
References
Howitt, D. L., & Cramer, D. (2004). The SAGE Dictionary of Statistics: A Practical Resource for Students in the Social Sciences (illustrated reprint ed.). New York: SAGE.
Ott, L., & Longnecker, M. ( 2010). An Introduction to Statistical Methods And Data Analysis (6 illustrated ed.). New York: Cengage Learning.
