a425_trend

# a425_trend - Time Series - Trends &amp; Seasonality APS 425...

This preview shows pages 1–6. Sign up to view the full content.

Time Series - Trends & Seasonality APS 425 - Advanced Managerial Data Analysis (c) Prof. G. William Schwert, 2002-2010 1 APS 425 – Winter 2010 Time Series Analysis: Trend & Seasonality Instructor: G William Schwer Instructor: G. William Schwert 585-275-2470 schwert@schwert.simon.rochester.edu Topics • Deterministic trends & seasonality • Stochastic trends & seasonality • Differencing & seasonal differencing

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Time Series - Trends & Seasonality APS 425 - Advanced Managerial Data Analysis (c) Prof. G. William Schwert, 2002-2010 2 Time Trends Linear Z t = a + b t + e t • do you really expect such a pattern to persist forever? • e.g., accumulated horse manure in the streets of Philadelphia in the late 18 th century exponential (linear in logs) log(Z t ) = a + b t + e t Z t = exp{a + b t + e t } In Eviews, time = @trend(K) creates a time trend equal to 0 in observation K 20 25 Linear Trend Example Actual Fitted 5 10 15 0 0 1 02 03 04 05 06 07 08 09 0 1 0 0
Time Series - Trends & Seasonality APS 425 - Advanced Managerial Data Analysis (c) Prof. G. William Schwert, 2002-2010 3 140 160 180 Exponential Trend Example Actual Fitted 40 60 80 100 120 0 20 0 1 02 03 04 05 06 07 08 09 0 1 0 0 Polynomial Trends Z t = a + b 1 t + b 2 t 2 + e t • parabola •i f b 2 < 0, then it eventually turns negative f b 2 > 0, then it eventually sky-rockets up • you can approximate most samples with polynomials, but you wouldn't expect to do well forecasting out-of- sample

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Time Series - Trends & Seasonality APS 425 - Advanced Managerial Data Analysis (c) Prof. G. William Schwert, 2002-2010 4 14 16 18 Quadratic Trend Example 4 6 8 10 12 Actual Fitted 0 2 0 1 02 03 04 05 06 07 08 09 0 1 0 0 Other Functions of Time Cosine or Sine represent fixed cycles S-curves (logistic) -- new product sales log((L - Z t )/Z t ) = a + b t + e t Z = L / {1 + exp[ a + b t + ]} Z t = L / {1 + exp[ a + b t + e t • L is the upper limit (asymptote) for sales • can iterate on this by trying different values to see which fits best
Time Series - Trends & Seasonality APS 425 - Advanced Managerial Data Analysis (c) Prof. G. William Schwert, 2002-2010 5 10000 12000 S-Curve Trend Example Actual Fitted 4000 6000 8000 0 2000 0 1 02 03 04 05 06 07 08 09 0 1 0 0 Other Types of Stationarity-inducing Transformations it •per capita • divide by population (if it is growing) •real • divide by price index (e.g., CPI) •market share • divide by industry sales

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## a425_trend - Time Series - Trends &amp; Seasonality APS 425...

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online