---
title: "sales"
output: pdf_document
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(data.table)
library(foreign)
sales <- data.table::fread('~/Dropbox/pkg.data/sales/raw/20142015 Comcast w SurveyQs Conf.csv')
spss <- foreign::read.spss('~/Dropbox/pkg.data/sales/raw/2014 2015 Comcast w 2015 Survey Qs Conf.sav')
```
# Sales project: Comcast Data
If I choose a goal level higher than one would predict then asymmetric information.
Here is asymmetric based on econometric modeling. Try to confirm or disconfirm the model.
Likelihood of achieving: 80\%
Likely_L4: Should be likelihood of baseline
Behavioral data: What they choose (likelihood): Survey data (likelihood of achieving each goal). Signals not only over confidence but asymmetric information.
Question of how hard they are going to work. If likelihood is 100\% but don't have to work that hard.
## 2014
1. Goal Selection
2. Performance Data
## 2015
1. Goal Selection
```{r}
pre.start <- which(names(sales)=='PreSurveyTaken')
pre.end <- which(names(sales)=='Gender')
pre.surveyFields <- names(sales)[pre.start:pre.end]
```
GoalLevel[2014,2015]
BaseObjective[2014, 2015]
GoalLevel/GoalLevel
2. Survey (Nothing about 2014 program)
a. all names between PreSurveyTaken and Dups `r print(pre.surveyFields)`
All performance fields start before *PreSurveyTaken*
3. Performance data
4. Post Survey (If there was a 2016 program what is the likelihood of achieving you would take goal level 1,2,3 next time )
a. PostEffort: The goal required a great amount of effort to achieve?
b. PostEffortSatsf: Are you satisfied with the amount of effort you put in?
b. PostGoalDiff: PostGoalDifficulty
d. PostEffort: Did the goal require a great amount of effort to achieve?
e. PostEffortBase[Lvl1, Lvl2, Lvl3]: How much effort would you need to put into to achieve
f. PostLikeBase[Lvl1, Lvl2, Lvl3: What is the likelihood you will achieve (assuming the effort you would put in)