The study shows that using certain statistical methods on data with a specific type of error can lead to inaccurate results. The researchers used a simulation to demonstrate that the standard method for analyzing this type of data can give misleading outcomes. They found that the problem doesn't get better with more data, but actually gets worse. They also discovered that this issue is common in U.S. commodity prices. The researchers suggest checking for this error type before using the standard method to analyze the data.