1. 100 to 1 in the Stock Market (4.5*) Jan
A detailed summary is available »here«
2. Fooled by Randomness (2.5*) 24-Feb
After reading the first few pages, I was shouting Ya, this is THE book that slams very many authors for their mistakes OR poisoning others with/without the author knowing it.
I enjoyed the book very much.
But why low ratings? I have expected the book to cover more of the randomness (as the title suggests), but roughly only 25% (only that much I could grab) covers the subject, that too in a superficial manner.
Rather the book talks about more poems, essays, historians, lengthy literature texts, etc.
Some of my grabbings are below.
The book emphasizes the possible parallel paths that didn’t occur, which end up as non-observed outcomes. A janitor may win a big lottery. If he/she plays the lottery one million times, there is not much scope for winning the lottery again. The other one million iterations are the parallel paths that didn’t occur and the janitor winning lottery is the random event that occurred. It is a random event, not the skill at work!
Same way if a million people are trading/investing in the stock market, for sure a few people will be more successful.
No wonder why we have Buffet and Munger.
Just like the lottery example above.
An another example is if 10k fund managers are starting their business today, they predict that stock A goes up/down that day.
Let’s say 50% of the fund managers vote for up and remaining for down.
The next day only the winning fund managers are allowed to play.
After the 10th day, there will be about 10 managers who have predicted correctly 10 times in a row!
Will you be trusting him with your life savings?
Here is an interesting discussion from the book:
A test of a disease presents a rate of 5% false positives.
The disease strikes 1/1000 of the population.
People are tested at random.
A patient’s test is positive.
What is the probability of the patient being stricken with the disease?
Most doctors answer 95%
Answer:
Consider out of 1000 patients one is expected to be affected with the disease.
Remaining 999 are healthy.
The test will identify about 50 with disease (it is 95% accurate).
\(Probablity = \dfrac{No.of. Afflicted}{No.of. True + false Positives} = \dfrac{1}{51} = 2\% \)
1000 observations won’t tell all swans are white - but observation of 1 black swan is enough to refute.