Estimating Nikkei Futures Fair Value from Dividends and OIS Rates

Summary

The document discusses practical inputs for a rough cost-to-carry estimate of a distant-expiry Nikkei 225 index future. Its proposed dividend input is a market-implied dividend strip, with the annual dividend amount translated into a continuous yield by dividing it by the current index level. This offers a way to incorporate expected dividends without modeling each payment separately.

For the financing input, the answer recommends constructing a TONA overnight index swap discount curve. It also suggests using professional market data services when available. The exchange provides no worked fair-value calculation, data series, explicit upper and lower bounds, or comparison between dividend strips and historical dividend forecasts. The advice is therefore a concise guide to input selection rather than a full pricing procedure; estimates will depend on market data access, dividend expectations, curve construction, and the assumptions used for a long-dated contract.

Key ideas

• A cost-to-carry estimate for an equity index future needs dividend and financing inputs.
• Market-implied dividend strips can provide expected dividends for the contract horizon.
• Annual dividends can be converted to a continuous yield relative to the current index level.
• A TONA overnight index swap curve can supply the financing or discount-rate input.
• The document offers input suggestions but no complete valuation or uncertainty bounds.

Tags

# Practically pricing index futures using cost to carry


# Practically pricing index futures using cost to carry












I am trying to compute a rough approximation for the theoretical price of a Nikkei 225 index future with a far-away expiry. I don't need much accuracy, just a very rough but reasonable upper and lower bound.

I understand the theory behind how to do this, but I need the data to actually do it. Can anyone point me in the right direction for what one would normally use for this?

I understand you could use dividend futures to approximate the dividends, or would it be better just to fit a simple model to the past dividends to approximate future dividends? Similarly, for the risk-free rate, where can I find data to compute the future expected interest rates to obtain a good approximation of the total cost to carry from the current to the expiry time?

## Answer by João (score 0)

https://quant.stackexchange.com/a/85223

Bloomberg's dividends forecast is actually pretty good, but turning each discrete dividend into a yield, seems like extra work for me. Instead you can use the market-implied dividend strip and convert the annual dividend amount into a continuous dividend yield

$$ q = \frac{D}{S_t} $$

For the RF you'll need to build the TONA OIS discount curve

If you have access to Bloomberg or LSEG is more than good, if not

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.