The correct answer is:

All of the alternatives are true.

Here's why:

• Use a small α when the series varies substantially: A smaller α gives more weight to past observations, which is helpful when the data is volatile, so it smooths out random fluctuations.
• Use a large α when the series has little random variability: A larger α gives more weight to recent observations, which is effective when the data has less random variation and trends can be captured more quickly.
• Use a value between 0 and 1: This is a fundamental rule of exponential smoothing; the smoothing constant α must always be between 0 and 1 to ensure a meaningful combination of previous and current observations.

Do you need more details on exponential smoothing or have any other questions?

Here are five related questions:

1. How does exponential smoothing differ from moving averages?
2. What happens when you set α to 1 or 0?
3. How do you select an optimal value of α for a given dataset?
4. What are the limitations of exponential smoothing in forecasting?
5. How does exponential smoothing handle seasonal variations?

Tip: When using exponential smoothing, test different α values to see which provides the most accurate forecasts based on past performance.