Since CFA Institute doesn't release the exact exam score we have to estimate the exact score with the provided ranges (<=50%, 51%-70%, >70%) for each topic. Although I don't really care about my exact score as long as I passed, it is a good educational project, which also shows the simplicity of Monte Carlo sampling. Please don't take this too serious, this was all done just for fun in about two hours.
We assume a discrete uniform distribution for each topic (i.e. each topic score within the range has the same probability, for example for the topic Economics which has a maximum of 24 Points, if you got 51%-70% in this topic, we assume 13, 14, 15 and 16 all have a probability of 25%). We produce samples of this distribution for each category, and the we sum up all categories and plot a histogram of the samples. This is also a good visual example of the Central Limit Theorem, which says the sum of any independent random variable will be normal distributed, as the number of summands goes to infinty. As we can see, the sum of 10 uniforms already looks approximately like a normal distribution.
- Install RStudio and download the source code
- Open
app.R - Install the required packages (
shiny,shinydashboard,ggplot2) - Click 'Run app'
This was done by Lukas Steinbrecher, if you have questions feel free to contact me.
This app is not affiliated with CFA Institute.
MIT