Financial Risk Manager Part 1

Pure time series models are not theoretically motivated. This means that they lack a solid theoretical foundation that explains why the current value of a variable should be related to its past values and to the values of a random error process. For instance, in the context of stock returns, it might be difficult to understand why the current value of a stock return should be related to its past values and to the values of a random error process without any theoretical backing. It would be more theoretically sound to explain return fluctuations using some macroeconomic variables that influence profitability, such as the state of the economy as a whole. This lack of theoretical motivation is a significant disadvantage of pure time series models when compared to structural models, which incorporate explanatory variables and are therefore more theoretically grounded.

Why other options are incorrect:

• B: Pure time series models can indeed produce forecasts. In fact, they are commonly used for forecasting purposes.
• C: Pure time series models can be used with various data frequencies, including high-frequency data.
• D: This statement is contradictory since pure time series models don't incorporate explanatory variables by definition, so there's no need to select them.