Probability distribution function for stock price given many parameters

External question — Quantitative Finance Stack Exchange Author: Aditya Tan Original post: https://quant.stackexchange.com/questions/71859 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. First of all, I am not in the US market. I am trying to find out if I can do probabilistic analyses of stock price movements using the buy and sell summaries. Kindly let me then explain my problem. My market is a little bit different from the US market; that is, we have access to daily transaction histories of different brokers (e.g. Robinhood). Some of these brokers are associated to big whales. Although orders may be split up into several different brokers (e.g. I may have accounts in both Robinhood and CS), I have a hypothesis that the pattern of buying and selling, vs the stock price will somehow show up. We have close to 100 brokers in the market, and I would like to identify two objectives: Any whales (really, conglomerates) who frequent certain brokers Could the purchase histories (accumulation or distribution) from any of these brokers be able to predict the price movement within a certain time frame (e.g 3 days, 1 week, 1 month)? Is the use of ordered probit, using binomial PDF, valid? So my input data looks something like: Any help is kindly much appreciated. Any follow up questions is also appreciated. Many thanks guys. Aditya