Quiz : Case Study

6. Analyze results - Graph the data from Table 4 and describe what your graph looks like (you do not have to submit a picture of the actual graph!).

2. Based on the scale of measurement for each variable listed below, which measure of central tendency is most appropriate for describing the data?

4. For the questions 1-3 we found out a couple different cost functions that could be useful for Delta when predicting salaries for 2003 and 2004. The multiple regressions may be most reliable considering it uses multiple variables, but the single R square is better than multiple regressions.

6. Although you are basically satisfied with the analysis thus far, you are concerned about the

Exercises 12.17, 12.21, and 12.43 require the use of the “Regression” function within the Data Analysis menu in Excel. Refer to Appendix E12 for instructions on using Excel for these exercises.

results from the model were both accurate and reliable. The multiple R and the R

4) Use exponential regression (or find a constant ratio) to determine an equation for the data.

Table 6.1.1 displays the matlab output of beta, standard error, t-statistic and p-value for the two independent variables during 10-year period. It is found that beta of X1 is 0.2750 which indicates there is a positive relationship between the utilities excess return and the healthcare excess return. This positive relationship is statistically significant as the p-value is close to 0 which is much less than the significance level of 5%. In addition, the standard error of X1 is 0.0300 which represents the average distance that the observed values fall from the regression line. This indicates that the model fits the data. In contrast, it is found that the material excess return is negatively

The business literature involving human capital shows that education influences an individual’s annual income. Combined, these may influence family size. With this in mind, what should the real estate builder be particularly concerned with when analyzing the multiple regression model?

14. See the attached “Regression Data I”. We are using the number of radios, TVs, and DVD players stocked to predict the profit, revenue, and cost for future periods. Next, run a model to predict the cost. Select all which apply.

In order to meet the demands of consumers while complying with the regulations, the auto manufacturer will invest money in converting crossover SUV’s and small light truck category vehicles to electric engines. Most manufacturers have this technology developed but simply need to further adapt it for these vehicles in order to improve each model’s mpg. In addition, the auto manufacturer will convert remaining light truck models to operate on a biodiesel engine. The manufacturer’s choice to focus on the light truck category rather than passenger car is due to “increasing mpg is much more valuable when the base is low than when it is relatively high” (Nichols & Datla, 2012, pp. 16, para. 4). For example, if an auto manufacturer’s passenger car average mpg is 40 and its light truck average mpg is 20 then the harmonic mean is 26.67. If the same manufacturer increased its passenger car average 5 miles per gallon to 45 mpg while

Use the sales data given below to determine: (a) the least squares trend line, and (b) the predicted value for 2000 sales.

This model was chosen because it experienced the most significant increase in adjusted R2 (up from 90.5%) and a decrease in AIC (down from 773.9) from the OLS model. The coefficients that were computed by the GWR tool and mapped (Figure 2) helped to demonstrate that each explanatory variable and its associated coefficient vary spatially in its predictive strength of the dependent variable. As we know, there is spatial autocorrelation and relationships in the data. This is not necessarily negative but it is important to capture the structure of the correlation in the model residues with explanatory variables. Until then, the model cannot necessarily be trusted (ESRI, 2016). However, the high level of significance of the p-value (0.0000) and the z-value (6.059441) indicate that the model can be trusted. The small p-value indicates that the coefficients are not zero and therefore the explanatory variables are statistically significant predictors of the behaviour of the dependent variable (ESRI, 2016). The small dataset is more troubling. Some future solutions for eliminating spatial autocorrelation include continuing to resample variables until there is no more statistically significant spatial autocorrelation (clustering). Unfortunately, that was not accomplished during the OLS regression without interfering with the ability of the GWR to run. The output of the OLS

For the model fit evaluation, the model has factor loading between 0.26 and 0.96, positive error variances and also reached level of highly statistically significance. Thus, passing the preliminary fit criteria. Meanwhile on the overall model fit criteria, as can be seen in table , the proposed model is achieving almost all of the ideal results. Only two indicators that the model didn’t achieved, the goodness of fit index (GFI) and Adjusted GFI (AGFI). However, the results are not too far from the ideal results from each measurements and Hooper et al. (2008) mentioned

We could also look at the model as a whole by looking at the F-Statistics and doing a F-test.