The graph may be used to estimate cholesterol level levels for doses between d_1 and d_2 but not outside this range. Doses outside the range may have unpredictable consequences, either reducing the cholesterol BELOW a safe level, or possibly even raising cholesterol by causing damage to the mechanism which raises cholesterol.

Predictions of future values may be extrapolate from past experience in the same way. For instance, when things are good, people imagine things will get better. They look at their increasing income for the past few years and imagine themselves on the escalator to prosperity. If income increased 5% each year for the past 5 years, they may think it will increase by 5% next year. In fact income is subject to the business cycle. Income goes up every year for several years and then down for a couple.

Don't extrapolate if you can avoid it. If you can't avoid it, leave a large margin for error. And bear in mind that the further you extrapolate outside the range of data for which you have a model, the larger the errors you may find yourself making.