But it seems contradicting. If one works like dividend theory (price gradual increase till ex div, and drops upon div payout) while another strictly follows movement of STI (go up 3%, same goes 3%), which is the better way to explain the corelation between STI and ES3?
The STI value is calculated based on
prices of the component stocks,
without considering their dividends. Since ES3 holds dividends from the component stocks, it will definitely differ from STI, depending on how much dividends it hold at that time.
(Check
page 20 of this STI document if you would like to know the exact calculation.)
The difference between ES3 and STI cannot be described as
tracking error. I would imagine/suspect that the fund manager will deduct the effect of dividends before calculating tracking error. We just have to accept/trust the tracking error reported by fund manager. I don't think there is a way for us to calculate on our own.