To the list of courses || GAT2018 || To the theme || Estonian

Exercise 3719. Points 2, theme: Model Application

Open exercise
Make a linear regression model for predicting examination marks from attached data using the following variables:
  • year (as a numerical variable),
  • student attendance in lectures,
  • student participation in practical lessons.
What mark is expected for a student in the year 2009 if the student did not visit lectures but participated in all practical lessons of his/her nominated student group and in half practical lessons of another group (participation in practical lessons = 150%)?
Add the software name you used.
Codes for marks are the following: A - 6, B - 5, C - 4, D - 3, E - 2. F - 1, MI - 0.

Data: GEO-AT_hinded-osalus.xls


In Statistica, select Multiple regression. The model parameters can be found in the result panel column B. After forming the model, select panel Residuals/assumptions/prediction, then Predict dependent variable, input values given in the exercise and press OK.

In the SDC find function Non-spatial regression and copy the data to the input cell. The dependent variable must be in the first column, the explanatory variables in the following columns. The SDC outputs the regression formula in results, the user has to add the legend (meaning of notations).

Input the given values to the cells in prediction panel in the SDC or use a calculator or Excel to derive the predicted value from the formula.Do not forget to add the regression formula and the legend of notation to your answer. An example of model legend is in the attached figure.

Log in to send your results and to see the expected answer and responses from other students.