Time Series Forecast with Only 2 Observations I have a number that I need to forecast its distribution over a period of time (5-7 years). There are currently 2 past observations, but I'd need to forecast the distribution of the rest. What options do I have given the limited data points? 
The distribution of the forecast should have a normal curve.
Moreover, I'm looking for insight concerning the programming aspects of this question. 
 A: Pretty much the only thing you can do in this situation is to take the average of the two points and use this as an expectation forecast. If you in addition want a normal distribution density forecast, then calculate the standard deviation of your two data points, and expect this as the future standard deviation.
Take this forecast with a humongous grain of salt.
Alternatives include using the last observation as your point forecast (the so-called "naive forecast"), or drawing a slope through your two data points and extrapolating this out. This last option is almost guaranteed to end in catastrophe, because it (a) assumes a deterministic and noise-free data generating process that (b) will not change in the future, but (c) you have a giant lever.

The best piece of advice I can give you is to invest in gathering more data, either a longer history, or extraneous data that can help you forecast your time series. This is definitely a better investment of your time than trying to find a method that will give you a meaningful forecast based on just two observations.
