Steps overall

1. Collect data
2. Prepare data for analysis
3. Perform statistical test to select variable
4. Select variable for modeling
5. Descriptive Analysis
6. Correlation
7. Prepare hypothessis to find target and independent variable
8. Correlation of target variable with all dependent variable wit p-value
9. Prepare AI/ML Model list
10. Build Model
11. Model Insights / Results
12. AI/ML Model API

Steps specific project

Here are the steps:

1. Develop a python tool to pull data from the FB ads
2. Prepare data for analysis
3. Perform descriptive analytics
4. Prepare Hypotheses and AI/ML Model for the objective
5. Build AI/ML model

• Training the model
• Testing the model
• Validation and Iteration of the model

1. Then Model insights will be visualization/delivered.

Output Structure

Interpret the Coefficients in Regression Models

How one interprets the coefficients in regression models will be a function of how the dependent (y) and independent (x) variables are measured.  In general, there are three main types of variables used in econometrics:  continuous variables, the natural log of continuous variables, and dummy variables.  In the examples below we will consider models with three independent variables:

x1i              a continuous variable

ln(x2i)         the natural log of a continuous variable

x3i              a dummy variable that equals 1 (if yes) and 0 (if no)

Listed below are three models.  In each case, the right-hand side variables are the same, but the dependent variables differ.  In each of these regressions, the dependent variable will be measured either as a continuous variable, the natural log or a dummy variable.  Define the following dependent variables:

y1i              a continuous variable

ln(y2i)         the natural log of a continuous variable

y3i              a dummy variable that equals 1 (if yes) and 0 (if no)

Below each model is text that describes how to interpret particular regression coefficients.

Model 1: y1i = β0 + x1iβ1 + ln(x2i)β2 + x3iβ3 + εi

β1 =∂y1i/∂x1i = a one unit change in x1 generates a β1 unit change in y1i

β2 =∂y1i/∂ln(x2i) = a 100% change in x2 generates a β2 change in y1i

β3 = the movement of x3i from 0 to 1 produces a β3 unit change in y1i

Model 2: ln(y2i) = β0 + x1iβ1 + ln(x2i)β2 + x3iβ3 + εi

β1 =∂ln(y2i)/∂x1i = a one unit change in x1 generates a 100*β1 percent change in y2i

β2 =∂ln(y1i)/∂ln(x2i) = a 100% change in x2 generates a 100*β2 percent change in y2i

β3 = the movement of x3i from 0 to 1 produced a 100*β3 percent change in y2i

Model 3: y3i = β0 + x1iβ1 + ln(x2i)β2 + x3iβ3 + εi

β1 =∂y3i/∂x1i = a one unit change in x1 generates a 100*β1 percentage point change in the probability y3i occurs

β2 =∂y3i/∂ln(x2i) = a 100% change in x2 generates a 100*β2 percentage point change in the probability y3i occurs

β3 = the movement of x3i from 0 to 1 produced a 100*β3 percentage point change in the probability that y3i occurs