FRM Part I · FRM Exam Part I · Nonstationary Time Series
A series grows at a roughly constant percentage rate. An analyst estimates ln(y_t) = 2.0 + 0.03t. Which statement about the model is correct?
The series grows by about 3% each period. In a log-linear trend model the slope on time is the approximate constant continuous growth rate of the series, so a slope of 0.03 means roughly 3% growth per period.
- AThe series increases by about 0.03 units each period
- BThe series increases by about 3% each periodCorrect
- CThe series has a unit root with drift of 0.03
- DThe series is covariance stationary around a constant mean
Explanation
A log-linear trend implies a constant proportional growth rate equal to the slope, so about 3% per period. A linear trend would give constant absolute change. A deterministic trend is not a unit root, and the mean changes over time so the series is not stationary.
Did you get it right without looking?
One question tells you little. A timed set on Nonstationary Time Series shows your real accuracy, how long you take and where you lose marks.
More Nonstationary Time Series questions
- A researcher fits a log-linear trend model ln(y_t) = 4.00 + 0.02t + e_t to quarterly revenue, where revenue is measured in millions of dolla…
- A log-linear trend model is estimated as ln(y_t) = 4.0 + 0.02t. What is the forecast of y at t = 50, closest to?
- A risk analyst runs a Dickey-Fuller test on a bond yield series by regressing the change in y_t on y_{t-1}. The estimated coefficient on y_{…
- Monthly log sales of a firm show a seasonal pattern in which the seasonal fluctuation is the same size each year, while a stochastic trend a…
- A log-linear trend model for annual earnings is estimated as ln(Y_t) = 3.00 + 0.05 t. The model predicts earnings at t = 10. Using the exact…
- A series follows Y_t = Y_{t-1} + e_t with e_t white noise. A researcher regresses Y_t on an independent random walk X_t, both of length 500,…