Invest Without Emotions
Summary: Mar NIFTY 18000 calls added 65,14,850 contracts while 17400 calls shed 44,21,500. On the Put side of the equation, the 17500 strike added 80,10,000 while the 16850’s shed 4,60,800.
Dotted lines indicated actual underlying volatility. Solid lines are IVs.
One percent:
Two percent:
One percent:
Two percent:
A lot of traders close out their positions by the end of the day. It certainly reduces the risk embedded in overnight moves (c2o) but that has been falling over the years. So has the o2c range. Perhaps it is an artifact of the bull market and reversion to the 2011-2013 range is imminent. Something to keep an eye on.
Code and charts on github.
First the Wikipedia definition: Treasury management (or treasury operations) includes management of an enterprise’s holdings, with the ultimate goal of managing the firm’s liquidity and mitigating its operational, financial and reputational risk. Treasury Management includes a firm’s collections, disbursements, concentration, investment and funding activities. In larger firms, it may also include trading in bonds, currencies, financial derivatives and the associated financial risk management.
For a bank, this means asset/liability management, hedging interest rate risk, managing reserve and capital requirements, etc. It is also something banks provide as a value added service for their clients.
Banks break out segment revenues that include revenue derived from treasury operations. However, they don’t carve out how much of it was proprietary trading. This is the average quarterly revenue from treasury operations since June 2014 (in Rs. Cr.) of major Indian banks:
In fact, for a few of them, revenue from treasury operations exceed from those from retail banking.
The revenue from treasury ops is a black box. If most of that revenue is derived from prop trading, then investors need better disclosure of the risks that were taken. Recently, YESBANK got pummeled on a UBS report that raised doubts over their exposure to stressed companies. As a retail investor, we really don’t have a clue about the risks involved in holding equity in what could turn out to be a hedge fund in bank’s clothing.
So far, we have focused on the NIFTY for selling butterflies (Part I, II, III.) How would this look on the BANKNIFTY?
First, let’s have a look at the 30-day rolling returns of the CNX BANK Index, from 2010 to the present, the whole population:
Median: 1.79%
2014-present:
Median: 2.21%
Beginning of 2015-present:
Median: flat
Currently, the index is around 18700. That means a 100-point move translates to 100/18700 ~0.5%
If we do the same back-test we did to the NIFTY, this is what we find:
As with the NIFTY, the trade makes money if you know how to cut your losses. However, when the trade is live, how do we know what the future volatility of the underlying is going to look like? Without risk-management, a short-call butterfly strategy will encounter out-sized losses that wipe out all prior incremental gains.
Our previous post discussed how a simple SMA On/Off Switch based tactical algo can be enhanced by a volatility metric. We generated significant alpha by following a simple rule:
But what if we trained a machine on the same data and allowed it to decide when to go long and short?
We fed an SVM our volatility metric and the percentage distance from 50-day SMA. A 5-year training set was used to predict the next year daily long/short. We will not delve into the details of how SVMs work, Wikipedia does a decent job introducing the concept.
To make it easier to compare, we plot the wealth-charts for the NIFTY and BANKNIFTY indices side-by-side.
The black line is the Machine Learning Long-Short Model and the blue line is buy-and-hold. NIFTY and BANKNIFTY since 2011:
NIFTY and BANKNIFTY since 2013:
Buy-and-hold has two big advantages over a trading strategy: transaction costs and tax treatment. Here is how the different strategies compare with buy and hold:
It appears that the ML(V + 50-DMA) Long Short strategy works better on the BankNifty than on the Nifty. The out-performance of the ML model on the BankNifty more than compensates for transaction costs and taxation.
The ML model outperformed the NIFTY by an average of 12% in the last 4-years and the BANKNIFTY by 94% in the same period. The out-performance on the BANKNIFTY is considerable enough to warrant further exploration.