Use Statistics and Probability to Make Financial Forecasts
To forecast your finances, you watch for trends, patterns, and relationships, determine the probability of these influencing a particular outcome, and use that to model your forecast.
For instance, if government indicators predict that the economy is going to grow by 4 percent next year and you’ve assessed a correlative relationship of index-predicted economic growth and sales in a ½ ratio, then you should predict that the economic growth will contribute to a 2 percent sales increase next year.
Does that mean that sales will increase 2 percent next year? Only if nothing else influences your sales at all, because other factors may influence sales to either make them higher or lower, but the economic growth will have a bit of a positive influence on your sales.
Based on consistent trends over each month of the last three years of a steady 1 percent monthly sales increase, you may predict that you’ll continue to see steady growth over the next several years, but with a 68 percent probability of slowed growth as you find patterns where sales slowed every fourth year.
Perhaps you couldn’t figure out what variables were influencing that slowed growth, but after calculating the probability of it, you were able to determine that your sales have a definite possibility of a temporary slow-down.
In the stock market, the two things that are most commonly used to predict movements are earnings and price. These two items as predictors is completely insane because both tend to be too volatile and too easily manipulated to be useful indicators.
So what are good indicators? There’s a joke that’s passed around by American economist Paul Samuelson, which says, The market has predicted eight of the last five recessions.
Another, somewhat more accurate indicator, is the yield on Treasury bonds. The yield on these bonds tends to increase and decrease in a generally similar way to national GDP but just two to four years sooner. Still, ratios such as price-to-earnings are quite popular for predicting stock market movements.
Calculate the Altman’s Z-score
An interesting case of statistical financial projections is the Altman’s Z-score. This calculation is 72 percent accurate in predicting that a corporation will file for bankruptcy within the next two years. While not spectacularly accurate (better models are now out there), the Altman’s Z-score is a very simple equation to use and is accurate enough to prove a point. Here’s how the equation works:
Z = 1.2T1 + 1.4T2 + 3.3T3 + 0.6T4 + 0.99T5
T1 = Working Capital/Total Assets
T2 = Retained Earnings/Total Assets
T3 = EBIT/Total Assets
T4 = Market Value of Equity/Total Liabilities
T5 = Net Sales/Total Assets
Risk score ratings:
>3 = As risky as eating soup while wearing water-wings (very low risk of bankruptcy)
1.81–2.99 = As risky as jumping off the high-dive in loose-fitting swim trunks (moderate risk of bankruptcy)
<1.80 = As risky as swimming with sharks after taking a meat-bath (high risk of bankruptcy)
Using statistics and probability takes several different variables (the components of the different financial metrics), weights them each by the amount that each is able to predict bankruptcy in a standard deviation, and then adds them together to give us something called a z-score (a measure of observed distance from the mean for a particular value).
Together, they’re 72 percent accurate in predicting whether or not a corporation will go bankrupt in the next 2 years.