Singapore Entrepreneurs Network & Startup Organization

As for the case of EMH, i read quite a few articles debunking the theory. One of the example is, if the market is truly efficient, nobody would outperform the market in the long run. The price will truly accurately depict the intrinsic value of the stock. Hence no point trying to find an 'undervalued' stock, as the price is already a precise reflection. Something along this line. I can;t recall all the details i read. Anyway you can find these articles quite easily at wallstraits.com

Carrotcake, i;m curious. Since you are into trading, what method do you use since you believe TA is useless? You mention arbitrage before, but don;t think it's available to us, small retail investors right?

 

 No apologies needed. we are all here to learn from each other

 Sorry but i will have to disagree with you on this. There are defintely many many quality postings, which i really benefitted and gained a lot from. Especially the older threads. Of course there are some silly threads, including 1 stupid question i posted once, but for sure, if you look closer and dig up the older threads, there are many gem postings there. Anyway i'm also a paid member of their premium site, and the quality of the postings alone is worth the investment. And no, i'm not advertising for them. Just that, among all the investment sites i been to, i benefitted most from Wallstraits, with their wide diversity of threads from people from all walks of life. Also i do admire that they dare to put money where their mouth is, investing in stocks that satisfy their screen in a real dollar portfolio, open to members of the paid site. Of course open to criticism as well.

Speculators aside, if EMH is valid, could Warrant Buffett and Peter Lynch achieve what they had achieved?  Outperforming the market for decades that is.

I also do know that the folks that came up with the EMH theory got a nobel prize for it. However, for me, i'm more inclined towards the side of a non-efficient market. Oh yes, before i forget, i read somewhere that in the book 'A Mathematician plays the stock market', EMH is true only if a majority believe it to be false. And vice versa. More food for thought.

I think EMH is like a grey area, with similarly strong for and against. There are many practises build on the assumption on EMH, and on those against it. I'm not an expert on this, so i shall stop here as far as this is concerned.

I used to be quite interested in trading. But now i realised investing for the long term suits my style more. But i won't rule out the occasional speculation.

CANSLIM is more like trading then investing. I read and study the charts purely for knowledge sake. I may apply the entry and exit part from CANSLIM into how i invest. Thats it. Buffett approach is the way for me.

But for speculation, CANSLIM should be good enough for me.

Well, the people in the CANSLIM SG forum swears by it. Maybe it may not work for you, but for them, it does. For different individuals, the goals, preference and risk appetite may differ, therefore different approaches are utilised. For these people, CANSLIM work perfect for them.

I don't get it, which assumptions do you find "not practical"?

Efficient market = traders entries no better than random. The recent model on London Stock Exchange (this was featured by Discovery Channel) failed to predict historical bid-ask spread until mathematicians assume traders are entering/exiting at random. (efficient). Which gave me the impression that EMH came about due to incidents like this.

I mean, if it was found to be more practical to assume a non-efficient market then EMH would be long gone from university syllabus like pesudosciences (10% brain theory, mindmapping etc).

Well, the entire field of financial engineering is based on the assumption that the market is efficient isn't it?

'Zero Intelligence' Explains Stock Markets
By Jennifer Viegas, Discovery News

Feb. 8, 2005 — Stock markets around the world are full of shouting, gesturing traders who presumably control the floor, but a new study indicates that stock markets, in many ways, operate as though buyers and sellers possess zero intelligence.

The new research does not suggest that buyers and sellers actually do possess such minimal smarts, but rather that inherent constraints within a stock market's fundamental structure exert a more powerful, predictable force than previously realized.

The findings could prevent stock market crashes. They also might lead to better ways of predicting stock market behavior, which could help to alleviate volatility and eventually could result in lower transaction costs.

The new research is outlined in a paper published in last week's Proceedings of the National Academy of Sciences.

The latest model of zero intelligence economics, developed by lead author J. Doyne Farmer, a physicist and instructor at the Santa Fe Institute, is startling. With near faultless precision it determines key characteristics of the London Stock Exchange.

"While many properties of markets have been claimed to be driven by the strategic behavior of agents in the past, our study casts down on this by suggesting an alternative explanation," Farmer told Discovery News.

Farmer explained that his model makes a few simple assumptions. The first is that traders place buy or sell orders at random, subject to constraints imposed by current prices. He does not consider information about current prices rational strategy, since most traders should have access to such "minimal intelligence," which Farmer and his team feel is low enough to be negligible.

In real markets, there are many different types of orders. Farmer says these can be broken down to two basic types: limit orders and market orders. A limit order does not cross the best price and so stays in the system until it does. But if an order to sell is lower than the highest price in the system, for example, it's executed immediately — a market order.

Economists group traders into bears or bulls. Farmer's model instead assumes that some traders will act with patience, while others will act with impatience. He links the patient traders to limit orders and the impatient ones to market orders.

Farmer and his colleagues created mathematical equations for each of these factors and then tested the resulting model on actual data from the London Stock Exchange. They plugged in information from 11 stocks that were active between Aug. 1, 1998 to April 30, 2000, a period that involved a whopping six million buy and sell orders.

The researchers chose the London Stock Exchange over other markets because it is fully automated and it is not cleared manually every few minutes, as the New York Stock Exchange is.

"We believe these results would also apply in the NYSE or NASDAQ markets, but the data to test this is much more expensive, and because the market structure is not as 'pure' the theory is not as directly applicable," Farmer said.

With 96 percent accuracy, the model predicted the bid-ask spread for the chosen London market stocks. The spread, which reflects what people are will to pay versus what others are willing to sell their stocks for, determines transaction costs.

Seventy-six percent of the time, the model accurately predicted the stocks' price diffusion rate, a figure used to measure price volatility and financial risk. Farmer and his team are not yet sure why that part of the model was less successful than the bid-ask component.

While not rated with a percentage point, the average market impact, which is the impact that trades have on prices, also was predicted by the model.

Farmer and his team say their findings can help both stock market traders and regulators.

"A practitioner who is worried about risk may find our explanation of volatility useful, for example, because it gives insight into why one stock is more volatile than another, or why some periods of time are more volatile than others," Farmer said. "Similarly, anyone who trades often is interested in the bid-ask spread."

He said regulators could use his data as a benchmark against which to measure the functioning of markets. The model also indicates that lower volatility is optimal, and can be kept in check by giving incentives to people who place limit orders, who are the liquidity providers, while charging those who place market orders, who are the liquidity demanders.

Blake LeBaron, a professor of international economics at the International Business School at Brandeis University, works on similar economic modeling. He said the new study was indeed successful at determining spreads and diffusion rates.

"Traders might be able to use this as a background model for thinking about new trading strategies," LeBaron told Discovery News. "They could use it as a kind of test bed for their strategies."

LeBaron added that not all stock markets, as of yet, can be predicted by models that assume zero intelligence. So far, persistence in volatility, direction of exchange rate movements, and trading volume issues have stumped economic model creators. In future, both LeBaron and Farmer hope to address those glitches.