What’s a Wainwright Map? |
| Wainwright maps serve several functions. First, they provide a snapshot of the forecasting model. The great majority of our models employ three or four input variables. Maps, however, being two-dimensional can allow for only two variables to be plotted. In some cases we can get around this restriction by combining two variables into one; for example, we can take inflation into account by multiplying a variable by changes in the prices of the precious metals. Naturally we choose the most predictive set of variables for use in a map.
|
 |
| Secondly, a map provides a reliable assessment of the historical performance of the model. This is not a back test of a specific formula; instead it is a real-time picture of how well the values of the input variables as they were at the time corresponded with the actual subsequent directional outcome a year later. We plot “tramlines” on each map to help with this determination. The area between the tramlines encompasses the immediate vicinity of the “0” or neutral line. Points that lie in this area represent outcomes for which the input data do not give a definitive forecast, and thus they are hard to predict directionally. Outside these lines, however, the results should be predominantly positive beyond the + tramline and negative beyond the – tramline. |
| Thirdly, a map is a forecasting tool. We have plotted the current inputs for the next twelve months for each developed market and marked this point with a star labeled “conditions as of date.” But users of these maps can do their own forecasting at any time.
|
| To do so, you need to calculate the simple change in the three-month Treasury bill rate from four years prior to date and locate the corresponding number on the y-axis. For the horizontal coordinate you need to consult the current Morgan Stanley Capital International month-end index of total return, determine the ratio between it and the U.S. index and plot this number at the appropriate place on the x-axis. These coordinates will reveal the direction of the forecast.
|
| If you want to get more specific and produce a numerical forecast, first draw in the neutral line parallel to and centered between the tramlines. Then draw a second line from the neutral line and at a 90-degree angle to the point defined by the coordinates for the input variables. This line is shown in each map. By using the tramlines as a guide, you can interpolate or extrapolate to get the numerical forecast.
|
