Quant? Quant? Kalman Filter and Baynesian for Finance all in one blog post?
What is a Quant? A Quant or sometimes Rocket Scientists is an information specialist who is able to use formulas from the space program, or deep math to create forecasting systems for Hedge Funds. An example might be the use of Kalman filters or Dynamic Bayesian Networks or Bayesian Belief Networks (yes that is real, for once I didn’t make it up), think about the movie “Margin Call” and if you haven’t seen it you might want to. Consider that you might get bored by set piece kinds of movies, the movie is all dialog, but it does talk about Quants, and that must be why you are reading this right?
Why see the movie? In one screen Jeremy Irons character states: “Explain what this does, as if you are speaking to a small child.” And the “Quant” explains the what the formula that the hedge fund is predicting.
To find out more about Kalman Filters and why you should care take a look at the paper and Excel example of how to do a Kalman filter using Excel, although the author doesn’t consider it a wise use of Excel, it because the author doesn’t make use of the full capability of Excel.
In any case this is a great example of a univariate Kalman Filter that you could use to create a paper on how advance engineering math is applied to financial markets. Now should you use this for your own investments? No. But take a look at it, you certainly could make use of the spreadsheet if you are in dire need of a discussion point for a math class and you stayed up late last night chasing after some desirable company, worked late, were sick (but not if you have been drinking, in that case you can’t use this example…just kidding, of course you can, also, aspirin helps out).
Now what is a Bayesian equation? Similar to the Kalman, the Kalman was designed to track objects, fast objects that were intended to break things you liked, and kill you. Bayesians were designed to model multiple stochastic processes, Kalman basically models the single track of an object. Big difference to me is the whole kill me component. You don’t use Bayesians in life or death situations.
Now if you are a financial kind of person, then the discussion abo the Kalman filter is something you WOULD use the cloud to implement. Use the Kalman filter “solvers” in Excel, generate the capability to connect your Excel program to the cloud and then hit the switch you can share the program with others for a subscription fee!
OR, OR, you would simply download Excel PowerPivot (free) and use your Excel in the cloud. Check it out here:
And here is why analytics is important, motivation rapper (at the end):