Paper: arXiv 2306.12439
Authors: Yuxia Liu, Qi Zhang, Wei Xiao, Tianguang Chu
Abstract
We propose a successive one-sided Hodrick-Prescott (SOHP) filter from multiple time scale decomposition perspective to derive trend estimate for a time series. The idea is to apply the one-sided HP (OHP) filter recursively on the updated cyclical component to extract the trend residual on multiple time scales, thereby to improve the trend estimate. To address the issue of optimization with a moving horizon as that of the SOHP filter, we present an incremental HP filtering algorithm, which greatly simplifies the involved inverse matrix operation and reduces the computational demand of the basic HP filtering. Actually, the new algorithm also applies effectively to other HP-type filters, especially for large-size or expanding data scenario. Numerical examples on real economic data show the better performance of the SOHP filter in comparison with other known HP-type filters.
Complexity vs Empirical Score
- Math Complexity: 8.0/10
- Empirical Rigor: 6.0/10 (capped from 6.5: no out-of-sample test; see penalty flags)
- Quadrant: Holy Grail — high math complexity, high empirical rigor (effective empirical rigor capped from 6.5)
Why this score: The paper is heavily mathematical, featuring matrix operations (Woodbury identity), derivations of incremental algorithms, and complexity analysis (O(l³) vs O(l²)). It also demonstrates empirical rigor by testing on real S&P 500 and Shanghai Composite data, comparing performance against established baselines (HP, bHP, OHP), and providing a visual time-consumption plot.
Research Flowchart
flowchart TD A["Research Goal: Improve Trend Estimation for<br>Nonlinear Economic Time Series"] --> B["Key Methodology:<br>Successive One-Sided HP (SOHP) Filter"] B --> C["Incremental Filtering Algorithm<br>(Addressing Moving Horizon Optimization)"] C --> D["Data Inputs:<br>Real Economic Time Series"] D --> E["Computational Process:<br>Recursive Application of OHP Filter<br>on Cyclical Components"] E --> F["Key Findings/Outcomes:<br>Better Performance vs. Other HP-Type Filters<br>Reduced Computational Demand"] F --> G["Conclusion: Effective for Large/Expanding Data"]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes a successive one-sided Hodrick‑Prescott (SOHP) filter and an incremental HP filtering algorithm to improve trend estimation on multiple time scales.
Source sentence
We propose a successive one-sided Hodrick-Prescott (SOHP) filter from multiple time scale decomposition perspective to derive trend estimate for a time series.
Assumes 5
- The OHP filter can be applied recursively on updated cyclical components to extract higher‑frequency trend residuals. model
Source sentence
applying the OHP filter recursively on updated cyclical components allows us to extract the leftover trend residuals of higher frequencies successively
- The tuning parameter η is set to 14400 as suggested by prior literature. statistical
Source sentence
with the tuning parameter η = 14400 in Eq. (2) as suggested by [12, 13]
- The logarithm of the stock price series is used as the input variable yt. data
Source sentence
Let yt be the logarithm of these stock data, we apply the four filters respectively to obtain trend estimates
- The incremental algorithm requires only a third‑order initial inverse matrix to generate inverses for arbitrary finite‑size data. model
Source sentence
with the only need of computing a third order initial inverse matrix once for all
Show 1 more
- The SOHP filter does not use any future data at any time (causal filtering). model
Source sentence
to involve any future data in processing at any time
Data
- S&P 500 index market-prices public
Source sentence
... consist of 885 observations including monthly data from January 1947 to September 2020
- Shanghai Composite Index (SHCI) market-prices public
Source sentence
279 observations including monthly data from July 1997 to September 2020
Costs
Not applicable
Finds 3
- The SOHP filter evidently improves trend estimates compared with HP, bHP, and OHP filters. performance
Source sentence
the SOHP filter evidently improves the trend estimates in comparison with the HP, bHP, and OHP filters
- The SI (Bayesian‑type information criterion) attains its minimum at n=4 for S&P 500 and n=3 for SHCI, indicating optimal iteration numbers. statistical
Source sentence
According to our numerical experiments, the SI in Eq. (12) assumes its minimum value at n = 4 and n = 3 for the S&P500 and SHCI respectively
- Mean and variance of the final cyclical component ct are small (e.g., mean 2.40e‑4, variance 2.70e‑3 for S&P 500). statistical
Source sentence
Table I shows the mean and the variance of the final ct, as well as the number of iterations and the SI value for the SOHP filter
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 15 related papers). Coverage is partial, so this is neither replication nor refutation.
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.