A Simple Bond Allocation Model v1

Example, Application & Back-Test

I have been looking for various ways to build strategic asset allocation models over the years, including bond models. The goal is to have a relatively simplistic rule set out and try to backtest it as far as possible. This post is part of a series where I present some models and strategies that I have found to be useful.

One such model is a bond allocation model, which I recreated with data going back to 1986 and updated for the most recent periods. In simple terms, the model looks at 30-year treasury yields, the Dow Jones Utility Index and Treasury bills in order to obtain signals. In this case, the decision will be to allocate to bonds or treasury bills. It’s either one or the other. In its original iteration, the model is spitting out astounding returns from 1943 until 1997. The results are quite impressive, with an annualised return of 8.72% while only showing a 3.2% maximum drawdown.

How has the model fared since then, and can it be potentially useful? Let’s have a look. There are some interesting results out of the engine room.