I need to have a few pages report. If it can fit in two pages, would be great.
I have the data for 7 types of hourly observations (24 of each) for 59 year period in the SAS format. All analysis and model development, have to be in SAS. Some calculations can be done in Excel.
The data is the weather observations from one station. I need to write on my approach to the developing the logistic regression model using this data, or the selcted predictors from this data. I have to explain in my report how and why I develop the steps in developing the predicting model (logistic regression).
The data is hourly, for each day, 24 observations for each type of the observation.
The predictor, the fog parameter, is already prepared as the dichotomic parameter, Dailyfog1, which equal to 1 for the day with fog observed (even if it was observed during one hour of the day) and this parameter equal to 0 if fog was not observed during the day.
I have already calculated the simple statisic for this parameter on monthly basis.
1. We need to o Select the months highly associated with daily fog events for development of the fog occurrence simulation model, and leave other months out from the analysis.
2. Have to present the reason why we select this month, and how we did the selection.
Also, I have to present the program which was used to make the selection, if such was used.
(I will provide you with my previous work. I already have the histogram of mean monthly fog occurrence).
3. Using the logistic regression built in SAS software to develop a fog occurrence simulation model.
o Evaluate the model performance using the measurements of logistic regression.
Reasons and Rationale are most important part in this research and the report.
--"include any key messages on what should be looked at in terms of developing a daily fog occurrence simulation model. The report should have specific information on regression methods selected, including the rationale on why the method(s) is(are) suitable for the dataset. In addition, the report should also include information on the expected model outputs or results following running the SAS procedures."
4. o Evaluate the model performance by analyzing 1) the relationships between fog occurrence frequency and probability derived from the logistic regression and 2) the relationships between the daily number of hours of fog occurrences and probability derived from the logistic regression. These evaluations can be displayed by tables or figures.
this is how the task is written by my supervisor.
As I need this project prepared for the end of Monday.
The report has to be in APA style, with short introduction, explaining methods , the approach, provide the reasons for it, have some first outputs and with Summary, with Conclusion.