Allocation3M

Machine learning and statistical data analysis have been increasingly used in work practices to improve business decision-making and increase work efficiency. Allocation3M combines three mathematical statistical models to help you analyze resource allocation issues in your work. Resources can be time, manpower, products, budget, etc.

Allocation3M contains four tools:

Growth Rate Data:
Add the growth rate data you want to analyze.
You can select any data for 12 consecutive cycles for analysis. For example, you can allocate service manpower based on the dealer’s monthly sales volume and seek the resource allocation ratio with the best sales growth rate. For example, you can allocate a procurement budget based on the monthly market prices of raw materials and seek the allocation ratio with the least price volatility.

Mean-Variance Model:
Seek the resource allocation ratio with the best expected growth rate.
The Mean-Variance model calculates the expected growth rate, volatility, and the growth rate per unit of volatility based on the past data of the allocation targets. The Mean-Variance model uses the Monte Carlo method to obtain the resource allocation ratio with the best expected growth rate and the lowest volatility.

Black–Litterman Model:
Modify the expected growth rate to calculate the resource allocation ratio.
The Black–Litterman model combines the user opinion, the Mean-Variance model and the Bayesian estimation method to calculate the resource allocation ratio with the best expected growth rate and lowest volatility.

Risk-Parity Model:
Make the contribution of each allocation data to the overall volatility the same.
The purpose of the Mean-Variance model and the Black–Litterman model is to optimize the expected growth rate, while the Risk-Parity model is to optimize the volatility. The Risk-Parity model uses Newton's method to calculate an approximate resource allocation ratio so that the volatility contribution of each data to the data set is consistent.