Futures, forwards, options, or, in general, derivatives, comprise a few examples of what comes to mind regarding the price of a security, commodity, or
currency in the future of the financial markets. You may recall when many
economists and analysts predicted the price of gold would be anywhere from $800 to
$1000 this year (2014). So far, it hasn't happened. But why? The reality is that most
analysts look at supply and demand to anticipate where the market is heading. The
question is whether the supply and demand analysis is consistent with market participants’
behavior and subsequent psychology behind it. What if investors are heavily betting on gold call
options with a strike price of $1200 or more for the entire year, and future
contracts of gold for this year are all above $1100? Which one do you really
trust? Those who invest on calls/puts/futures? Or those who analyze historical data, predict the future demands, but have no skin in the game? Perhaps you listen to both. But what if
their estimates are not even close?
What about analyzing the future data and not
historical data? Historical data has already been processed in peoples’ mind, and they
are expressing their forward-thinking views via what they paid for option of the underlying security. What if we could create a simple model to read those investors’ minds? Let’s plot
a graph that shows not only the price curve from past up to now, but also illustrates
what happens from now on. To plot the graph, we need data points. One way to get them is to call any potential investors of the predicting security to tell us what they think of tomorrow's stock price, the price day after, and so on for next 365 days, and then we average them out. But that's far too time-consuming and impractical. So let's cheat. Instead of getting data from what people think, we get it by looking at what they thought.
Taking the current price of their latest call/put options for each option's expiration date, we draw a smooth curve through the data points. Now that we have dates, how do we calculate the data points by looking at options’ prices?
For each expiration date (our data point) we multiply the number
of call options (open interests) by last price/bid price and multiply them by their
respective strike price. We calculate this for each strike and we add them up
for all strikes. We do the same thing for
put options. We add these two together and divide them by their total weight. For
calculating the weight, we basically add the number of call options times their respective
prices to the number of put options multiplied by their respective prices.
Now let’s demonstrate this with some real data. I've specifically chosen a large capital growth company (Tesla) with a very high volatility average of 10% or more on any given day. I also chose a day when the market is experiencing the highest volatility for the last couple of years. NASDAQ went down almost 18% in one month (NDAQ: 40.79 on March 7th to 34.77 on April 11th). Many traders are still anticipating the worst is yet to come, and they are expecting this market to behave like 1987 or the 2000 dot com bubble. Except in this case there is no mass sell-off, or bad economic news. In fact the volume is pretty much the same or even lower than usual. But there are simply more sellers than buyers. In my opinion, this is just the Fed's tapering effect (but that's another blog post). TSLA
closed at $216.93 on April 9, 2014. If you look at the option chain (you can check it in Google finance - very convenient), there are 11 option expiration dates available beginning April 11, 2014 and ending Jan 15, 2016. There are your data points. Strike prices start from $155 to $340 for April 11, 2014.
Total number of calls Open Interests X strike price X bid price = 17864568.73
Total number of puts Open Interests X strike price X bid price = 18967387.68
Total number of calls X bid price = 84386.23Total number of puts Open Interests X strike price X bid price = 18967387.68
Total number of puts X bid price = 81709.16
Estimate Price for TSLA on April 11, 2014 Expiration Date=
(17864568.73 + 18967387.68) / (84386.23+ 81709.16) = 221.75