The relation between fund flows and returns in the Australian ETF industry is assessed in this study. Daily data from 43 equity ETFs over the period 2019-2023 are used. The main research objective is to accentuate whether past returns can predict future fund flows and in what way and vice versa. According to the results of the applied regression analysis, past flows and past returns can predict to some extent concurrent flows. This is also the case about concurrent returns. However, in both cases, the results are not unanimous and depend on the specification of the applied regression model.

This study compares return and volatility performance of exchange-traded funds (ETFs) with high-ESG (Environment, Social, and Governance) rating vs. low-ESG rating. The paper also examines time-series data predictability by identifying their positive dependence and volatility asymmetry properties, and examines the performance of two combinations of short-memory models i.e., autoregressive moving average and exponential generalized autoregressive conditional heteroskedasticity (ARMA-EGARCH); autoregressive moving average and asymmetric power autoregressive conditional heteroskedasticity (ARMA-APARCH) and two long-memory models, autoregressive moving average and fractionally integrated exponential generalized autoregressive conditional heteroskedasticity (ARFIMA-FIGARCH); and autoregressive fractionally-integrated moving average and asymmetric power autoregressive conditional heteroskedasticity (ARFIMA-APARCH). The study found that low-ESG rating ETFs on average have slightly significant higher returns and also lower volatility compared to their high-ESG rating counterparts. Evidence of asymmetric volatility properties are also present on both high-ESG and low-ESG rating ETFs returns. The study also observed that for both high-ESG and low-ESG rating ETFs denote a stationarity, but non-invertible process in their returns. Results can provide fresh understanding in the topic of leverage effects and volatility that can open future research channels to academicians.

We present two approaches to forecasting parameters in the SABR model. The first approach is the vector autoregressive moving-average model (VARMA) for the time series of the in-sample calibrated parameters, and the second is based on machine learning techniques called epsilon-support vector regression (ε-SVR). Using daily data of S&P 500 ETF option prices from January 1, 2014, to December 31, 2018, we first calibrate the daily values of the model parameters from the training sample, then conduct out-of-sample forecasting of parameters and pricing of options. Both approaches produce good fits between the forecasted and calibrated parameters for out-of-sample dates. A comparison study shows that using forecasted parameters as inputs, the SABR model generates better pricing results than assuming constant parameters or using lag parameters. We also discuss the market conditions under which one approach outperforms the other.