What is the predicted outcome when applying a linear regression model to a non-linear relationship?

When a linear regression model is applied to a non-linear relationship, the predictions generated are likely to be biased and may mislead. This is because linear regression assumes a linear relationship between the independent and dependent variables. In cases where the actual relationship is non-linear, using a linear model can result in systematic errors in predictions, as the model fails to capture the true pattern of the data.

For example, if the true relationship is quadratic or exponential, fitting a straight line to the data will not adequately reflect how the output changes with varying inputs. As a result, you might see predictions that trend in the opposite direction of the actual data or miss key fluctuations that are indicative of the real relationship. This misrepresentation can lead to incorrect conclusions and decisions based on the analysis.

In contrast, the other options suggest scenarios that do not align with the inherent limitations of applying a linear model to a non-linear scenario. For example, a linear model cannot yield perfectly curved predictions or establish a relationship with high confidence when the underlying association is not linear.