## Portfolio Performance of the SDR and Reserve Currencies: Tests Using the ArCH Methodology

_IMF Working Papers, February 1, 1993_

## Source details

**Canonical URL:** [Portfolio Performance of the SDR and Reserve Currencies: Tests Using the ArCH Methodology](https://www.imf.org/en/publications/wp/issues/2016/12/30/portfolio-performance-of-the-sdr-and-reserve-currencies-tests-using-the-arch-methodology-1017)

## Other formats

- [Markdown version](/en/publications/wp/issues/2016/12/30/portfolio-performance-of-the-sdr-and-reserve-currencies-tests-using-the-arch-methodology-1017/index.md)
- [Structured JSON version](/en/publications/wp/issues/2016/12/30/portfolio-performance-of-the-sdr-and-reserve-currencies-tests-using-the-arch-methodology-1017/index.json)
- [Bundle manifest](/en/publications/wp/issues/2016/12/30/portfolio-performance-of-the-sdr-and-reserve-currencies-tests-using-the-arch-methodology-1017/bundle-manifest.json)

## Bibliographic details
- Authors: Tugrul Temel, Michael G. Papaioannou
- Published: February 1, 1993
- Series: IMF Working Papers
- DOI: https://doi.org/10.5089/9781451842630.001

---

### Summary
- Study tests the assumption that the variability of returns on various currency assets is time invariant.
- Uses autoregressive conditional heteroskedastic (ARCH) models to evaluate time variation in variances of returns.
- Data: weekly data for the period February 1982 to December 1991 for major reserve currencies, including the SDR.
- Main finding: variances of returns do vary over time (i.e., they do not exhibit stationarity).
- ARCH models that specify changing variances are superior to models that assume constant variance.
- Error from incorrectly assuming stationarity is smaller with the SDR than with any other national currency in the portfolio to be managed.

### Methodology
- Modeling approach: autoregressive conditional heteroskedastic (ARCH) models and comparisons with constant-variance specifications.
- Data frequency and period: weekly data; February 1982 to December 1991.
- Asset coverage: major reserve currencies, including the SDR.

### Key Findings
- Evidence that variances of returns vary over time (nonstationary variance behavior).
- ARCH-family models outperform constant-variance models for currency-return data in the sample period.
- Mis-specifying constant variability leads to suboptimal currency-asset allocations and inaccurate measures of portfolio riskiness.
- Relative robustness: the SDR exhibits smaller error from an incorrect stationarity assumption than any single national currency in the examined portfolio.

### Policy and Practical Implications
- International investors and central banks should account for time-varying volatility when constructing and evaluating foreign exchange portfolios.
- Using ARCH or ARCH-family specifications can improve risk measurement and asset-allocation decisions for fixed-income currency portfolios.
- Inclusion of the SDR in reserve portfolios may reduce error arising from incorrect stationarity assumptions relative to relying solely on national currencies.

---


_Source: https://www.imf.org/en/publications/wp/issues/2016/12/30/portfolio-performance-of-the-sdr-and-reserve-currencies-tests-using-the-arch-methodology-1017_