George H. K. Wang

Modeling nonlinear dynamics of daily futures price changes

The purpose of this article is to characterize linear and nonlinear serial dependence in daily futures price changes. The daily prices of four futures are included in this study: (i) SP (ii) Japanese yen; (iii) Deutsche mark; and (iv) Eurodollar. Our major empirical findings are: (i) Based on the results of nonlinearity tests (that is, the BDS, the Q-super-2, and the TAR‐F tests), we found all futures price changes contain nonlinearity in the series; (ii) a GARCH model can explain the source of nonlinearity for three out of four series; (iii) a threshold autoregressive model and autoregressive volatility model can adequately represent nonlinear dynamics of SP and (iv) deterministic chaos is not evident in the scaled residuals from the nonlinear time series models. Hence we favor a statistical time series approach to represent the data‐generating mechanism of futures price changes.

Determinants of the ARM share of national and regional lending

We model the ARM share of mortgage lending and provide several unique contributions to the mortgage choice literature. First, we motivate the use of the price spread between fixed- and adjustable-rate credit as a regressor by constraining the effect of FRM and ARM prices to be symmetric and show that the data support this restriction. Second, our data span a far longer time period (six years) than previous research. Third, we estimate separate share equations by region, allowing us to contrast geographic variation in ARM shares. Fourth, we examine the effect of convertible ARMs—which became prevalent in mid-1987—on overall ARM lending.

The Split of the S&P 500 Futures Contract: Effects on Liquidity and Market Dynamics

The Chicago Mercantile Exchange reduced the size of its S&P 500 futures contract when it reduced the multiplier from 500 to 250 and increased the minimum tick from 0.05 to 0.10 on November 3, 1997. This is a rare major change in a very successful contracts specifications. We analyze effects of this change on liquidity and market dynamics in both a univariate and a multivariate context. The main contribution of this study is the use of multiple intervention analysis with various dynamic response functions to examine the effects of the split while taking into account several other major market events surrounding it. A multivariate analysis is also used to test the impact of the split using a structural model of liquidity and market dynamics. Empirical findings offer limited support for the hypotheses that smaller contract size resulted in smoother trading, and that more public customers trade the S&P 500 futures contract following its split. We observe a reduction in the average transaction size as well as a temporary narrowing of the bid-ask spreads, but no significant change in volatility that can be attributed to the split. We do not find any significant and lasting impact on other liquidity and market variables.

A Test of Integration and Cointegration of Commercial Mortgage Rates

Little empirical work examines the extent to which commercial mortgage markets are integrated into broader capital markets. We use time series data on commercial mortgage yields and yields on comparable-maturity Treasury securities to identify a long-run cointegrating relationship between the two yield series. Our empirical evidence suggest that, while the yield on commercial mortgage is cointegrated with that on comparable-maturity Treasury securities, the cointegrating relationship is far less than that found between the yield on residential mortgage rates and that on comparable-maturity Treasury securities during 1980–1990 time period. However, our results also show that the spate of commercial mortgage securitization that began in early 1991 may have been a market-integrating force and caused the commercial mortgage market to become more integrated into broader capital markets. Indeed, our results suggest that changes in capital market rates are now much more rapidly reflected in commercial mortgage rates than in the 1980–1990 time period, although there is a lag.

Temporal Relationships among Adjustable-Rate Mortgage Indexes

This article investigates the linkage among six ARM indexes during the 1978–1989 period. Grangers direct causality test is used to examine their relationship within a rolling regression framework. The nonstationary properties of each index and selected pairs of indexes are investigated by using the unit root and cointegration tests. The empirical results confirmed their relationship has changed over this period and short-term rates lead the eleventh district cost-of-funds index. The implications of the empirical results from the perspectives of borrowers (ARM choice), lenders (pricing), and investors (security valuation) are also discussed.

A monte carlo investigation of the BDS statistic

This paper investigates the finite sample distribution of the BDS statistic based on Monte Carlo experiments and analyzes the results of the experiments using response surface techniques. The major findings of the study are as follows: (1) the finite sample distribution of the BDS statistic of independent and identically distributed (IID) time series is closely approximated by the standard normal; (2) when the IID data comes from uniform or bimodal normal distribution, the distribution of the BDS statistic exhibits departure from the standard normal even when the sample size is large; (3) the BDS statistic has high power to detect both linear and nonlinear serial dependence when the sample size is approximately 1,000 and when the serial dependence in the time series is either moderate or strong; (4) the BDS statistic has relatively low power when the sample size is small and when the serial dependence in the time series is weak; finally, (5) the BDS statistic can be used as a residual diagnostic tool to c...

Seasonal Variation in Cost-of-Funds at Thrift Institutions

The purpose of this note is threefold. First, in addition to the well-known seasonal pattern to the eleventh-district cost-of-funds (COF), we document a twelve-month seasonal in the national median COF. Second, we demonstrate that the cause of seasonality in each of these COF series is due to the maladjustment of length-of-month effects. In particular, the eleventh-district COF is biased upwards in relatively short months while the national median is biased downward. Third, we show that the popular partial adjustment model for modeling the COF is misspecified. Copyright American Real Estate and Urban Economics Association.