Applied Statistical Model
Topic of Interest The daily phenomena in our everyday lives that we encounter that is, to determine the places and choices of food we consume, and restrictions on them based on our daily expenses. Title The ability to bring a change to our daily expenses (savings) through our food consumption Brief background of this topic . We are often pressured by the same mind boggling question, that is “What are we to consume for lunch/dinner daily? . This often arises and in order to keep up with our expenses according to our affordability, we can weigh in the option to consume food hat are not only healthy, but are cheap, mass produced and inexpensive. Our goal is to create a restaurant to cater to the needs of all consumers let it be young or old. Objectives : To determine the factors affecting the monthly expenditures on food.
To analyze the extent to which individual spend their expenses on food to meet their individual needs daily Analyze the prices of the array of food selection in certain shops to determine what meal can be consumed To create a general awareness about our food cost and it shows the importance of examining food costs per meal To ensure the safety and quality of food Discuss the research methodology First and foremost, we took the first step towards starting this assignment is that we collected our data through a survey research method. We have one dependent variable that is, monthly expenses on food and 1 1 other variables supporting it.
The other variables are age, monthly allowance, maximum amount willing to spend on food, number of meal consumption in a day, and the types of food ( Western, Chinese, Indian, Malay, Italian, Japanese, others). We drafted out several simple questions to be handed out to 35 random students and we collected our data once we received enough information. Later on, we computed all the data given using the Microsoft Excel software. Moving on, using the information that we conducted through the survey, we interpreted the data. We also used the SPAS software to transfer all our tabulated data.
Using the SPAS software, we managed to check the linearity of the model, the significance model of the data, relationships in the data, as well as to analyses large datasets . Besides that, the correlation of dependent and independent variables of the data can be tested. When the assumptions of the linear regression are not met, then data transformations has to be performed. A software that is named after ‘R’ is used to transform the Y to a new variable to meet the assumptions. To transform the data, we use box Cox after that. Box Cox with parameter, is used to power transform and to determine the new variable, Y.
Model Adequacy checking Normal Probability Plot Since all of the points lie along the straight line, the error term is normally distributed, hence normality assumption is met. Since there is no obvious pattern to be found in the residual plot, equality of error variance assumption is met. A) No outlier in the data because all standardized residuals is within В±3. ) No transformation is needed for the model because both the assumptions are met. Analysis Scatter Plot -Positive but weak relationship between the Y and the age of consumers. -Positive and strong linear relationship between the Y and the monthly allowance of the consumers. Positive and weak relationship between Y and the amount willing to spend on food a day -Positive and fairly strong linear relationship between Y and the number of meal consumptions a day Variables Entered Variables Removed Method dimensions Preference on Other Food, Maximum Amount Willing to Spend on food, Preference on Chinese Food, Age of Consumers, Preference on Malay Food, Preference on Japanese Food, Number of Meal Consumption a day, Preference on Italian Food, Preference on Western Food, Preference on Indian Food, Monthly Allowance . Enter a.