Regression Analysis Assignment Help
Regression Analysis Assignment Help Data and its interpretation playing a major role in research and development. A well-arranged data can revel at most information about the core problem. Answer is always hidden inside the data. From various techniques we can explore its inner nature and the relevancy about our problem. In business world at most problem can be handle able by its data analysis and survey. Initially people prefer that a better survey can validate the existence of the problem and then by various mathematical and statistical techniques can be applied to get the answer of our main problem. Using predictive analysis of the data one can decide in the market is what to do and what not in a real time mode. From the beginning of the business getting closer to the customer is one of major target of the promotional activity f the business and if one can get success in this then he can lead in his business very easily. But how one can go closer to each of his customer is a tuff task but in this digitally advanced era it can be done in a single click. Before this click a rigorous analysis of the data is involved. To set the trend of a business one can get a direction from the interest of their customers and decide in a real time that what their next strategy is. This prediction analysis can easily decide.
- It identify the hidden pattern in the current trends of the customers
- Using various models we can decide that which customer is in risk.
- By the measurement of likeliness of the customer we can decide the futuristic goal of the business.
There is a problem we discussed here is about to know the trends of online shopping between male and female for a particular online shopping portal. This data can reveals so much information from their statistical analysis. Further we will show this particular analysis. From an online survey we got the purchase value of consecutive 15 year between male and female. The problem is to find the trends of material on the portal and find the access of material with popularity among the people according to their gender.
Database for the analysis of our problem discussed in section 1.
The above data is taken from the website [A]. We can see that the quantification of a physical problem is basic need for the proper analysis. Always we try to get the survey or the primary data in numerals. Because of statistical measures we can easily predict the hypothesis and verify its validity for further work. It gives the strength for our assumption. The quantification can be done in various ways, for example we can get a survey by just filling some quantities between 1 to 10 or we can get the data in a continuous manner also. So mainly they are of two categorization one is discrete and another is continuous. In the problem discussed in 1 the data we got is a discrete type data.
|variables||Type of data||Statistical comments|
|A_P_Female||Numarals||It is quantified data and on an average the total purchages i.e. mean of the data = 63.800$|
|A_p_Male||Numarals||It is also quantified and average purchage of male on this website is of 22.45$.|
|in $||IN $|
In the report we can see the mean and standard deviation of two distribution of purchasing amount between male and females. Let X be the data of purchasing price for male and Y be the data of purchasing price of female.
Now we follow these steps to perform t- test
- Place the raw data X in column x, and Y in column y
- Calculate mean for both.
- Calculate deviation for both group by subtracting each score from it’s group mean and squaring it and put these in the columns “(x-Mx)2” and “(y-My)2″
- Sum the squared deviation scores for each group
- Calculate S2for each group
- Set up formula
- Calculate t
- Check to see if t is statistically significant on probability table with df= N-2 and p < .05 (N = total number of scores)
Regression Analysis Assignment Help
We can also analyze the data by regression analysis of these two variables. Regression and correlation analysis are on of the major statistical tool to get the connection between two distributions. In business world we have to know the connection of two data so that we can predict another one by knowing some of its variable. Correlation coefficient [B] is a measure to find the existence of relationship between two random distributions. Because before any analysis we need to know the degree of relationship between two distributions. It give the strength to our prediction analysis. In regression analysis we can find the exact relationship between these two variables.
Following are the steps to find regression analysis.
- state the hypothesis
- now the null hypothesis
- collection of data
- analyze the distribution
- Access the relationship between variables
- Calculate the regression equation of the data
- Test the statistical significance
- Accept or reject the hypothesis.
Now we are following the steps of t test and following are the result we obtained
|Paired Samples Statistics|
|Mean||N||Std. Deviation||Std. Error Mean|
|Pair 1||in $||63.8000||15||19.20268||4.95811|
|Paired Samples Correlations|
|Pair 1||in $ & IN $||15||.908||.000|
|Paired Samples Test|
|Paired Differences||t||df||Sig. (2-tailed)|
|Mean||Std. Deviation||Std. Error Mean||95% Confidence Interval of the Difference|
|Pair 1||in $ – IN $||41.34026||15.27651||3.94438||32.88041||49.80011||10.481||14||.000|
Conclusion: The t test is a test to find the validity of our hypothesis [C]. In spss we can calculate the t value of these two distributions and find that the standard deviation in the data of men is less than the standard deviation of female case. So we can conclude that the fluctuation or variability of data is large in the case of female. In generally we can say that men are consistent customers on this portal, female are also consistent but less than men. According to average values of both of the distribution we can see that mean of female data is very huge with respect to men’s data. So we can conclude that female community giving a large profit to this company, so it is an interesting conclusion for the company that female community not so consistent but they providing a large sum of profit. For promotional activity company should concentrate on the special customer who showing interest in the product of this online shopping portal. Since female community data have large average values and large standard deviation so we can think that female are not consistent but they provide more profit so company should concentrate on them and enhance their performance in online market.
Now the result of regression analysis is as follows:
|Model||Sum of Squares||df||Mean Square||F||Sig.|
|a. Dependent Variable: in $|
|b. Predictors: (Constant), IN $|
|Model||Unstandardized Coefficients||Standardized Coefficients||t||Sig.|
|a. Dependent Variable: in $|
The regression line will be
X= 3.917 Y – 24.182
Now we know that regression analysis based on the correlation coefficient between X and Y. In our calculation of coefficients for regression analysis we found that they are highly correlated and the coefficient f correlation is 0.908. This is higlypositive[D]. Correlation coefficient lies between -1 and +1 .in this case it is very close to 1 so we can say they are positively correlated. Two positively correlated data are directly proportionate. i.e. we can say that if one data is incising then another will also increase or if one is decrease then other will also decrease. In our case there are two variable of purchases amount of male and female community on a specific online marketing shop. X and Y are two variables showing these two distributions. From the value of correlation coefficient we can say that if female are going for more shopping then male will also going for shopping. This connection lead use to find the dependent and independent variable. In our case Y is called independent variable and X is called dependent variable. Since X depends on Y and they are positively correlated so we can predict the value off X by the above regression relation. In predictive analysis regression line will fit a curve or line which can help us to predict any value of X for a given value of Y. Here Y is the purchasing amount by female customers and X is the purchasing amount by male customers. So in any case we can predict the purchasing amount by male customer if we know the purchasing amount by female customers. It helps the company to decide their next production slots of materials for male or female. In the case of inventory Management Company need a pre determine strategy to produce more stuff according to their demand and if we know or predict the demand in male community by the data of female purchasing data then we can easily manage the inventory.
[A] Bouwmeester, S., Sijtsma, K., &Vermunt, J. K. (2004). Latent class regression analysis for describing cognitive developmental phenomena: An application to transitive reasoning. European Journal of Developmental Psychology, 1(1), 67-86.
[B]Hesterberg, T., Moore, D. S., Monaghan, S., Clipson, A., & Epstein, R. (2005). Bootstrap methods and permutation tests. Introduction to the Practice of Statistics, 5, 1–70.
[C]De Winter, J. C. F., &Dodou, D. (2010). Five-point Likert items: t test versus Mann-Whitney-Wilcoxon. Practical Assessment, Research & Evaluation, 15, 11.
[D]Rost, D. H. (1991). Effect strength vs. statistical significance: A warning against the danger of small samples: A comment on Gefferth and Herskovits’s article “Leisure activities as predictors of giftedness”. European Journal for High Ability, 2, 236–243.