|Chennai||Rs. 25020.00 (0.81%)|
|Mumbai||Rs. 25890.00 (0.98%)|
|Delhi||Rs. 25200.00 (-0.2%)|
|Kolkata||Rs. 25480.00 (1.03%)|
|Kerala||Rs. 24800.00 (0.61%)|
|Bangalore||Rs. 25000.00 (0.81%)|
|Hyderabad||Rs. 25080.00 (1.09%)|
Christmas is about making wishes. However, some wishes are not reasonable. For example, wishing for a universal database is wishful thinking, as such a database would need a lot of server space. I am not speaking of a smart stock market database but a database for every organic data from the stock markets to the stars. Scientists know the limitations connected with cosmic data. What the data astronomers are capturing, on occasion, comes from the past, from a few millions years back in time.
In continuation with our last article on turbulence, predictive science is about a few aspects, stolid databases, smart query systems and artificial intelligence, which can anticipate patterns in data or trends in data. Now artificial intelligence regarding database systems might seem science fiction and universal databases might be wishful thinking. This leaves us with just smart query systems built as funnels under a deluge of information.
Google is like this prototypical funnel under a deluge of information. It has its limitations. I was looking for an American jazz singer I heard a few years back and I just can’t find her with all the current search tools. The search is not cognitive or semantic yet. Google could not help me reach the singer because I needed related search parameters. What was her age? Was she an African-American? What was her net worth? Suddenly, something so relevant for me got lost in the deluge of information. The problem with search is lack of smart catalogued databases, which can understand each other. Only when databases are able to understand each other can data come alive and make search smarter.
Why should databases understand each other? If cause and effect can stretch to unrelated areas and Robert Prechter, Robert Folsom and Yale Hirsch could connect social trends, hemline lengths and presidential cycles to stock markets; behavioural finance could prove that there is no economics without psychology and fractal mathematicians could illustrate that nature was geometrical, then why can’t databases from diverse areas like psychology, economics, biology and stock markets talk to each other? Why can’t the way we search for an American jazz singer be the same manner in which we look for trading ideas? Why can’t intuitive become objective? They could, if databases had a common communicating language.
Even stock market systems are like database query systems. Trading systems are based on two broad approaches, as a trade signal in one asset (buy or sell signal on gold) vs. a trade signal in a group (best buy among 30 stocks). Though one could adopt a discretionary approach to stock picking, a database query can be adopted for both approaches.
Why can’t we have a database query for stock market investing? We can’t query because we don’t have smart databases to query - databases which contain fundamental, statistical, technical, sentimental and a host of other data elements. If we had a database like this, what could we query? We could have numerous queries. For example: What are the less-than-10 price to earnings multiple CNX 100 stocks, which are overbought over the monthly time-frame, are relatively outperforming the Sensex and the Dow, are outperformers in their respective sectors, have broken multi-week resistance and are in positive territory performing multi-week and multi-month cycles, have a volatility lower than 20 per cent annualised and are creating a positive trend on Twitter.
When we can run such queries, we mix fundamental, technical, statistical and sentimental data for a single query. This is when a database creates fusion and a market analyst ceases to be fundamental, technical or behavioural expert - rather turns into a data miner.
To test our talking databases, we ran a query on CNX 100 stocks. Which stocks were above multi-week resistance levels, were positive on performance seasonal cycles, had a positive multi-week price trend and were in the bottom among the relative underperformers in CNX100? Only six made the list. These are Sesa Goa, Axis Bank, Maruti Suzuki, ICICI Bank, LIC Housing and Aditya Birla Nuvo. We expect these stocks to deliver in 2013. And, if they do, the same database querying approach can be used to find the American jazz singer with a new search approach.
The author is CMT and Founder, Orpheus CAPITALS, a global alternative research firm