Random Matrix Theory

As discussed earlier, due to entanglement of financial markets with diverse world wide and tiny scales of people, considering methods which are able to evaluate behaviors in ‘temporal scale-wise networks’ during investment horizons are vital. These ‘temporal scale-wise networks’ contain leaders, followers and neutral stocks. This classification reveal how a certain stock behaves in comparison with collective behavior of market. This approach not only helps to better portfolio management, but also helps for better decision making in stock selection in markets’ turning points. Also, when market index is exposed to crash, in spite of acceptable fundamental aspects, stocks which are more entangled in collective phenomena of the market, may experience crash too. So, in bear markets, it would be better to be more separated from market collective behavior. Meanwhile, comparing two stocks which one has huge number of shares (and huge market cap), and the other one has small number of shares ( little market cap), they, definitely, will show different behaviors in different market’s movements. Besides these, it does not make sense that any co-behavior in the markets, yields us to significant correlation. To be certain, the results will be compared with the state of random situation (lack of any information). In the case of significant difference among these, these co-movements mean correlation.