OTHER REGRESSION
For a comprehensive overview of the effect that variables have reported mortality due to cancer, I considered appropriate to set various regression statistics, to see how to change variable changes the output of the regression. So, besides the variables mentioned in the previous chapter I have introduced two new variables, "distance" and "central". They are not real "new" variables because the latter is originated by the variable considered in the previous chapter "Coeff. Total ". I thought it appropriate, therefore, consider both the summary variable that the two original separately.
The following table shows the regressions are more significant for a greater understanding of the final output.
For a comprehensive overview of the effect that variables have reported mortality due to cancer, I considered appropriate to set various regression statistics, to see how to change variable changes the output of the regression. So, besides the variables mentioned in the previous chapter I have introduced two new variables, "distance" and "central". They are not real "new" variables because the latter is originated by the variable considered in the previous chapter "Coeff. Total ". I thought it appropriate, therefore, consider both the summary variable that the two original separately.
The following table shows the regressions are more significant for a greater understanding of the final output.
6.1 Before regression
For the first analysis I have considered the two main variables of the effects of radioactivity to see how the presence of nuclear power reactors or deposits in the province and the closest distance to the adjacent province can affect mortality [1] .
From this analysis it is clear, noting the extent of '= 0.1692, that the variables do not explain well the variability of the dependent variable. Evaluating the coefficients of regression certainly distance has a negative sign because it is inversely proportional to the impact of radiation on cancer mortality decreases as the distance increases the effect of the radioactivity man.
the plant's presence in the region certainly has a relationship with the output but does not affect much on it, it shows the t-statistic very close to zero.
6.2 seconds
In the second regression analysis, I used as an independent variable the coefficient value resulting from the two variables listed above, to have a real effect in each province, for two variables to consider separately might be an understatement.
The output of this regression is noted that the effect of the presence of radioactive material on the Italian territory is not particularly relevant to the determination of the effect on the dependent variable. It actually has a little higher than shown above, not explicit, therefore, a particular dependence on mortality. This analysis was intended to verify if the synthesis of the old variables in a single variable could eliminate any effects arising from discursive analysis of the two separate. But it was found that is not the reason that leads us not to accept as the most reliable, such as explanatory variables in mortality from cancer.
6.3 Third regression
In this analysis I wanted to analyze, as in the main regression, all variables taken into account, but I replaced the variable summarizing the total coefficient with the two generations, to see if the summary in one variable had created distortions in 'output.
From the regression did not suggest major differences with the main regression, synonymous with a good summary in a single variable but not a great dependence of the dependent variable from two variables (distance and power). Analyzing the t-statistic of the four variables we notice a slight scaling for the variables "income" and "elderly" in favor of the remaining variables, but certainly not so substantial as to change the final decision.
6.4 Fourth regression
In this regression I thought it appropriate to analyze the effects of radioactivity, measured by a single factor, with the variable of income, to see more effects of the combination of variables taken into account.
This analysis shows a high dependence on income, showing a very high value of the t-statistic. Unlike the value of the t-statistic, the coefficient for the variable total, there would not reject a hypothesis nothing.
6.5 Fifth
regression analysis I considered the latter as it can change the output analysis of the total coefficient approached the variable "old" and because the presence of radioactivity and the number of elderly in the Italian territory affect the output of the regression.
from this final regression, we can confirm the findings derived from the first regression: coefficient of radioactivity in the course of the analysis, the value of the t-statistic higher, but certainly the same statistics for the number of elderly is much more important. This analysis, the significance dell’ , è certamente la più significativa dopo l’analisi principale. Pertanto tali variabili riescono a spiegare gran parte dell’output della regressione totale e evidenziando la relativa importanza che ha la presenza degli anziani in ogni provincia come determinante della mortalità per tumore.
6.5 Sintesi regressioni
Al termine di tale lavoro ho ritenuto opportuno sintetizzare i dati ottenuti nelle precedenti regressioni in un'unica tabella utilizzando, dopo ogni regressione, i seguenti comandi di Stata:
estimates store first
estimates store second
estimates store third
estimates store fourth
estimates Store Fifth
and finally enclosing the previous commands in a table through:
Estimates table first second third fourth fifth
The table thus obtained can be observed at the end of this work in Appendix 2.
[1] The results of the regression, as for later, they were listed and summarized at the end of this work in Appendix 1.
For the first analysis I have considered the two main variables of the effects of radioactivity to see how the presence of nuclear power reactors or deposits in the province and the closest distance to the adjacent province can affect mortality [1] .
From this analysis it is clear, noting the extent of '= 0.1692, that the variables do not explain well the variability of the dependent variable. Evaluating the coefficients of regression certainly distance has a negative sign because it is inversely proportional to the impact of radiation on cancer mortality decreases as the distance increases the effect of the radioactivity man.
the plant's presence in the region certainly has a relationship with the output but does not affect much on it, it shows the t-statistic very close to zero.
6.2 seconds
In the second regression analysis, I used as an independent variable the coefficient value resulting from the two variables listed above, to have a real effect in each province, for two variables to consider separately might be an understatement.
The output of this regression is noted that the effect of the presence of radioactive material on the Italian territory is not particularly relevant to the determination of the effect on the dependent variable. It actually has a little higher than shown above, not explicit, therefore, a particular dependence on mortality. This analysis was intended to verify if the synthesis of the old variables in a single variable could eliminate any effects arising from discursive analysis of the two separate. But it was found that is not the reason that leads us not to accept as the most reliable, such as explanatory variables in mortality from cancer.
6.3 Third regression
In this analysis I wanted to analyze, as in the main regression, all variables taken into account, but I replaced the variable summarizing the total coefficient with the two generations, to see if the summary in one variable had created distortions in 'output.
From the regression did not suggest major differences with the main regression, synonymous with a good summary in a single variable but not a great dependence of the dependent variable from two variables (distance and power). Analyzing the t-statistic of the four variables we notice a slight scaling for the variables "income" and "elderly" in favor of the remaining variables, but certainly not so substantial as to change the final decision.
6.4 Fourth regression
In this regression I thought it appropriate to analyze the effects of radioactivity, measured by a single factor, with the variable of income, to see more effects of the combination of variables taken into account.
This analysis shows a high dependence on income, showing a very high value of the t-statistic. Unlike the value of the t-statistic, the coefficient for the variable total, there would not reject a hypothesis nothing.
6.5 Fifth
regression analysis I considered the latter as it can change the output analysis of the total coefficient approached the variable "old" and because the presence of radioactivity and the number of elderly in the Italian territory affect the output of the regression.
from this final regression, we can confirm the findings derived from the first regression: coefficient of radioactivity in the course of the analysis, the value of the t-statistic higher, but certainly the same statistics for the number of elderly is much more important. This analysis, the significance dell’ , è certamente la più significativa dopo l’analisi principale. Pertanto tali variabili riescono a spiegare gran parte dell’output della regressione totale e evidenziando la relativa importanza che ha la presenza degli anziani in ogni provincia come determinante della mortalità per tumore.
6.5 Sintesi regressioni
Al termine di tale lavoro ho ritenuto opportuno sintetizzare i dati ottenuti nelle precedenti regressioni in un'unica tabella utilizzando, dopo ogni regressione, i seguenti comandi di Stata:
estimates store first
estimates store second
estimates store third
estimates store fourth
estimates Store Fifth
and finally enclosing the previous commands in a table through:
Estimates table first second third fourth fifth
The table thus obtained can be observed at the end of this work in Appendix 2.
[1] The results of the regression, as for later, they were listed and summarized at the end of this work in Appendix 1.
0 comments:
Post a Comment