Plz help me! I truly do not know how to do.
The link for data is show as here : https://docs.google.com/spreadsheets/d/1y0oyFKnv7VTDCuWgPDdbvuk4AbAIn4oFZyE4Y2ZJHP0/edit?usp=sharing
A policy analyst for a local school board wanted to determine what relationships between income and agg'egate level of education might be used to encourage students to stay in school. Although there were potential problems with interpreting relationships based on aggregate data, she decided to begin with data 'om the 2016 Census. She collected data for the almost. 276 census tracts in OttawaGatineau, before compiling a dataset with the following variables: CensusT Pop 1 5+ P_hsgrad P_trades P_collcert P_1u1ivdipl P_bachdegr P_meddent P_masters Medlnc Angnc Medlnc' FITSI identifying code for the census tract the ntunber of adults aged 15 and over with earned income the proportion of adults with high school graduation the proportion of adults with qualications in a trade the proportion of adults with a college certicate the proportion of adults with a university diploma (no demee} the proportion of adults with a bachelor's degree the properties of adults with a medical. dental or related degree the proportion of adults with a masters degree the median employment income for individuals above 15 years the average employment income for individuals above l5 years the median employment income. with missing values the ts or predicted values rm: the model in part (f) Note that each proportion tracks the relative number of individuals whose highest level of education is as indicated and the categories are mutually exclusive. The data is in the le ottawagatineau.xlsx. (a) (1)) Plot the average incomes against the median incomes, or draw a boxplot or histogram of the average incomes. What two words would best describe the shape of income distributions in general and of average incomes in particular? Perform a multiple regression analysis using the seven educational variables as predictor variables and the median income (Metnc) as the response variable. Are there any problems with nnilticollinearity'? The VIF values are given below for each variable. Predictor VIF P_univdip| 1.1 P_hsgrad 4.1 P_bachdegr 8 P_trades 7.1 P_meddent 2 P_co||cert 2.6 P_masters 9.1 (c) For the regession model in (b), graph the residuals against the tted values and comment on whether the linear regression model assumptions are warranted. If you are using Minitab 17, you should plot the standardized residuals; otherwise, plotting the residuals is ne. ((1) Regress the P_bachdegr variable against the other six educational variables. \"that is the relationship between the Rsquare from this model and the V113 of the P_bachdegr variable from the model in (b)? (e) The Medlnc" variable copies the data from the Medlne variable, but a missing value code (*) has been inserted for a number of census tracts. Examine the Medlnc\" data and describe the nature of these census tracts (hint: look at the standardized residual values for the \"unusual observations\" from the regression model in part (b).) The remaining questions pertain to regression models based on the Medlnc* variable and not the original Medlnc variable. The elimination of these observations means that subsequent models may not predict well the median incomes for these unusual census tracts. (E! (s) (h) (i) (i) (k) (1) Re-estimate the multiple regression model using Medlnc" as the new response variable. What changes do you notice, comparing the model in (f) with the model in (b)? Plot the residuals against the tted values. Do you see any other problems with the model assumptions? If you are using Minitab 17, you should plot the standardized residuals; otherwise, plotting the residuals is ne. Calculate the correlation coefcient between the tted values (see FHSl in worksheet) and the Medlnc" variable. Show the relationship between this correlation coefcient and the value of R2. Perforur an F-test for the overall usefuhiess of the model, using the 1% level of signicance. \\Vhat do you conclude? Using the model developed in part (1), test the marginal usefulness or importance of the P_baehdegr variable, given the other variables in the model, using a 1% level of signicance. What do you conclude? If the proportion of adults with a bachelor's degree were to increase by 0.1 in a set. of census tracts (that is. from 0.1 to 0.2 or from 0.3 to 0.4, as the case may be). assuming the other predictor variables remain constant, what is the estimated average increase in the median incomes for these census tracts? [Give an estimate using a 99% condence level.) Would you conclude that a university degree is benecial in terms of increasing aggregate incomes? (1n) (n) (0) (p) Regress Medlnc* against only the P_bachdegr variable and nd the estimated slope of the regression line. Is the coefcient of the P_bachdegr variable in the simple regression model consistent with the coe'icient of the same variable in the multiple regression model? Explain briey why they might differ. Use the model developed in part (f) to calculate a 99% prediction inteival for the actual median income in census tract 907.00. In Minitab, select \"Predict\" under \"Regression\". Show manually how the standard error for the prediction interval is calculated using the standard error for the condence interval and the standard error (s8 = vMSE ) of the estimated regression model. Explain why you would not expect the prediction interval to cover the actual median income for this census tract. Finally, re-estimate the multiple regression model, but this time drop the variables that are the least useful. You can also use the \"Best Subsets\" option in Minitab to nd the best model. Explain whether this nal model is better than the model from part (i)
