Singapore Twitter Map

Some interesting features immediately come to light here. For instance the two huge vacant holes on the plot correspond to the two large national parks in Singapore. Concentration of tweets in other places also brings to light roads and other transportation conduits.

I also did some analysis around tweets by hour and day of week and kind of reverse geo-coded popular locations to figure out where people were tweeting from on weekends. The result of that exercise gave this interesting tag cloud

Popular Places on the weekend

I’ve put all this together in one gorgeous infographic which you can see here.
PS: Many Thanks to good pal Jayaram for helping me put the map together in R

Of course, investing in a  good mutual fund through a Systematic Investment Plan (SIP) is a no-brainer and I believe the corner stone of every portfolio (especially if you do not have the time to micro-manage your portfolio) should start there. However, I also believe it’s possible to augment returns from mutual funds through investing in stocks directly or through derivatives.

Which brings me back to the fundamental question – is there a fairly “sound” system for getting into the stock market and exiting profitably? From the little I know about the field, most academic research has been done in the “value investing” school of thought i.e. investing in stocks that are trading at less than their intrinsic values.

So I decided to investigate this further by building a “paper” portfolio of value stocks. I chose Benjamin Graham’s Net Current Asset Value (NCAV) per share as the criterion to filter stocks.  The NCAV per share is calculated as per the formula below:

What the NCAV per share effectively indicates, is a “realistic” cash value that may be available to the shareholder once all liabilities have been met.  Graham recommends buying stock that trades at approximately 67% or less of the NCAV per share. A study by the State University of New York has shown that investors could have earned an average return of 29.4% over a 20 year period, by picking stocks by this method and holding them for a year.
Once the filter criterion was decided on, I needed a universe of stocks to choose from. For the universe of stocks to choose from, I restricted myself to the stocks that compose the BSE SMALL CAP index (which also served as my benchmark). I then ran a series of scripts that I wrote in Ruby to pull the data needed to calculate NCAV from Yahoo Finance. Once I had all the data in a spreadsheet, I calculated the NCAV for all the stocks.
While Graham recommends only picking up stocks that trade at values significantly below their NCAV, I chose to go with stocks that were trading at prices that were at the NCAV or below it. Using this filter, I got a portfolio of about 17 stocks. The next step in my opinion should have been either of the following:

• Use the Markowitz optimum portfolio theory and allocate a fixed amount of capital among the stocks chosen so as to minimize portfolio risk OR
• Do further fundamental analysis (which most investment “gurus” recommend) and pick only those that are “fundamentally” strong

However, since I was pressed for time – I took the easy way out.

I just created a “paper” portfolio containing 100 shares of each stock. Slightly “unscientific” I know, but bear with me.
So with this small little NCAV portfolio – I began tracking the performance of the portfolio. And the results have been interesting, to say the least.

NCAV Portfolio Performance

The portfolio has significantly outperformed the benchmark over the last month. The returns from the benchmark BSE SMALL CAP index have been about -2.14% while the portfolio has returned about +9.09%.
While I agree that the time period is probably a bit short for significant analysis, I think the results warrant further investigation. Over the next few months, I will actively monitor my little test portfolio and update this blog. Lets see how it goes.

1. The APIs are Google’s property. It can do as it wants with them.
2. Most of the people who are crying out against this haven’t contributed in any form to either the API development or maintenance.
3. Almost all the depreciated APIs have replacements.
4. Quite a few of the APIs were still under API Labs – which means that they were experimental.
5. The Termination clause in the Terms & Conditions (which I assume most people expecting to run a business / project based on a API would read) clearly states that Google may terminate the API if, I quote:
   

   the provision of the Services to you by Google is, in Google’s opinion, no longer commercially viable.

6. The Warranty clause in the Terms & Conditions further states that:
   

   IN PARTICULAR, GOOGLE, ITS SUBSIDIARIES AND AFFILIATES, AND ITS LICENSORS DO NOT REPRESENT OR WARRANT TO YOU THAT:

   1. YOUR USE OF THE SERVICES WILL MEET YOUR REQUIREMENTS,
   2. YOUR USE OF THE SERVICES WILL BE UNINTERRUPTED, TIMELY, SECURE OR FREE FROM ERROR,
   3. ANY INFORMATION OBTAINED BY YOU AS A RESULT OF YOUR USE OF THE SERVICES WILL BE ACCURATE OR RELIABLE, AND
   4. THAT DEFECTS IN THE OPERATION OR FUNCTIONALITY OF ANY SOFTWARE PROVIDED TO YOU AS PART OF THE SERVICES WILL BE CORRECTED.

(Note the emphasis on the ALL CAPS text)
So, given all this, did all those complainers really expect to keep making money off Google’s efforts and data in perpetuity? And is basing a business model on a API, that clearly states that it might vanish overnight, really such a smart idea?

Google also drew a lot of flak for pulling the plug on Translate citing API abuse as the reason. A lot of people felt that, given how “smart” Google was, they should have found some other way round the abuse problem. My take on this is that, people give Google some credit. Given how “smart” they are I’m sure they must have explored all avenues before coming to that drastic conclusion.

I think in the end, its the developer community that needs to shape up and stop having unrealistic expectations of experimental APIs and be ready to change legacy code when API’s depreciate or stop using them. And Businesses need to realize that depending on other companies for core functionality is never a smart move. I’ve already been involved this year with cases that show how both depending on API’s and building missing functionality using API’s can seriously boomerang.

Thats my 10 cents on this issue. What’s your take on it?

Read the rest of this piece of analysis below…

Tedx Mumbai

I first installed one of the most popular plug-ins called AVH-Amazon. This is one of the most extensive plug-ins available for WordPress, so I figured that it ought to do the job. However, after installation I learned that due to a change in Amazon policy I would need to register at Amazon Web services and get a secret API key. The plug-in author didn’t provide one, as the key was going to be used by Amazon to monitor API usage and bill customers for use of those parts of its API that it charged a service fee for. Naturally the plug-in author didn’t want to risk it with his own API key, due to potential for misuse.

So faced with this roadblock and not wanting to go the extra mile and get an API key, I started looking for alternatives. I remembered that Amazon provided a RSS feed for a users Wishlist items. So I thought, what if I managed to extract the information from the feed itself ? As it is, I just wanted the item title, image and link to the item page, all of which were there in feed. Some tinkering in PHP and a few minutes later, I managed to parse the feed and show the items from the Wishlist. Voilà !! Instant Wishlist.

While I am happy with the result so far, I was wondering if there are any other good plug-ins that would let me do this a bit more efficiently. Any ideas ?

The very first project available for download is Loupe, the rails+flickr image gallery. This is the same gallery that I run my photography site on. So please visit the site, and download and try the scripts. Any feedback would of course be welcome.

But I digress, so back to Drupal. I took to Drupal since it seemed to be easier to set up and use. However in hindsight I think I was mistaken.

Initially setting up the system was quite easy. Theme and module installation was also quite simple. However, once the initial work was done the deficiencies started becoming apparent. For starters, the first thing I noticed was that there was no rich text editor. While this is more of a personal preference and I understand that some people might not want to use one, I was quite surprised that I had to actually install a module manually to get it working. Then also there were a lot of little things like no way to turn off breadcrumbs without physically editing the theme files, and no way to turn off the display of the author in a blog posts etc.

On the plus side, it’s blocks and taxonomy features are quite powerful and I really liked the flexibility that these features gave. But then again, on the negative side documentation on specific things that might be useful to a developer like a database structure and relationship between tables etc was not really easily found. I’m sure that the documentation might exist somewhere, just that I wasn’t able to find it. At the end of the day, I had to try and figure out things from the DB for some simple things I wanted to fetch and display. Though it does have a great API, with a nice DB abstraction layer, all this doesn’t really help if you don’t know what tables to query, now does it ?
To summarize, Drupal is  a powerful, clean and extensible CMS however if you want to very customized feature set for you site and don’t want to get your hands dirty in code then maybe Drupal isn’t for you.

Technorati Tags: Drupal, CMS