This project uses an exponential regression model to analyze the relationship between Previous Spending (independent variable) and Revenue Growth (dependent variable) for a specific data set.
Based on an initial analysis of the data, it appears that the relationship between Previous Spending and Revenue Growth is non-linear. Therefore, an exponential regression model was chosen as a potential model for the data.
The exponential regression model used in this project assumes that the relationship between the dependent variable (Revenue Growth) and the independent variable (Previous Spending) can be described by the following formula:
Revenue_Growth = a * exp(b * Previous_Spending)
where a and b are parameters to be estimated from the data. The log transformation is used to linearize the model:
The log transformation is used to linearize the model:
log(Revenue_Growth) = log(a) + b * Previous_Spending
The regression coefficients estimated by the model are in the log scale, so we need to transform them back to the original scale by taking the exponent of the coefficients.
The data set used in this analysis consists of Previous Spending and Revenue Growth data for a specific set of customers. The data was loaded from an Excel file and prepared for analysis.
The relationship between Previous Spending (independent variable) and Revenue Growth (dependent variable) is non-linear, and that an exponential regression model may be a better fit for the data than a linear regression model.
The exponential regression model assumes that the majority of the points are concentrated near the beginning of the axis. This suggests that there is a strong relationship between Previous Spending and Revenue Growth at lower levels of spending, but this relationship weakens or becomes less clear as spending levels increase, which is characteristic of an exponential relationship. It's important to note that other types of non-linear relationships may also exist between the two variables, and it's possible that a different type of non-linear regression model may be a better fit for the data. Therefore, it's important to consider multiple types of non-linear regression models, and choose the model that best fits the data based on criteria such as goodness of fit measures and interpretability.
This project was created by Aleksandar Dimitrov and is licensed under the MIT License. If you have any questions or comments, feel free to contact me at alexi.zein@gmail.com.