Question 1: Using dataset “Display_data.csv”
We have 30 days of data that tell us how much we spent, how many
clicks, impressions and transactions we got, whether or not a display
campaign was running, as well as our revenue, click-through-rate and
conversion rate.
- Describe your hypotheses.
- If we were to make a hypothesis, the hypothesis test that would be
best in this situation is a test of proportion. The alternative would be
that the variables have a significant relationship with the money spent
on ads to the revenue earned. The null would be that the opposite is
true, this is that the variables have no relationship with the revenue
earned.
- Report your findings using either an output file or a RMarkdown
file.
- I have created a report in R and I will attach at the end of this
notebook.
- Say we want to predict revenue based on “spend”. Please start a
simple regression model with one predictor, “spend.” Explain your
outcome and make managerial recommendations.
- Based on the simple regression model I was able to create in my
analysis using a R Markdown model, I was able to find that the results
concluded with the variable “spend” being positive and statistically
significant, suggesting that an increase in spending will be positively
associated with an increase in revenue. A managerial recommendation that
could be made off of these results is that budget should be allocated
strategically in order to maximize the revenue earned, based on these
past spending patterns.
- Please start a multiple regression model with two predictors
“spend”, and “display.” Explain your outcome and make managerial
recommendations.
- Based on the multiple regression model I was able to create in my
analysis, I was able to find that the results concluded with the
predictors “spend” and “display” each influencing revenue independently.
Because of their own influences, the managerial recommendation in this
instance would be to optimize the spending and the campaign efforts to
maximize the revenue potential. This is the best decision considering
the joint impact of these two variables on revenue.
Question 2: Using dataset “ab_testing1.csv”
A local retailer in Bakersfield wanted to test consumers’ reactions
toward their advertising campaigns. You were assigned to help this
retailer run the campaigns, perform marketing analysis, and make
recommendations. To do so, you randomly divided 30 targeted consumers
into three groups (Ads = 0, control group or no ads exposure; Ads =1,
version 1 of the ads; Ads =2, version 2 of the ads). The goal of the
experiment is to see which ads campaign lead to more product sales. You
will be testing the relationship between advertising and product
purchase (purchase) using regression analysis.
- Describe your hypotheses.
- If we were to make a hypothesis, the hypothesis test that would be
best in this situation is a test of proportion. The alternative would be
that there is a significant relationship between at least one version of
the ads and product purchase. The null would be that there is no
significant relationship between advertising exposure and product
purchase.
- Report your findings using either an output file or a RMarkdown
file.
- I have created a report in R and I will attach at the end of this
notebook.
- Explain your outcome and make managerial recommendations.
- Based on the liner regression model I was able to create in my
analysis, I was able to find the p-value which is less than 5%, which
means the null hypothesis can be rejected in favor of the alternative,
which means that it is true that there is a statistically significant
relationship between those who received advertising exposure and
purchasing products, versus no ad exposure. Based on the min and the max
going from negative to positive, this indicates that Ads1 was the group
that earned more and therefore was more effective.
References
https://rpubs.com/drosales15/midterm - Question 1
https://rpubs.com/drosales15/midterm-pt2 - Question
2