Source: Qilehui, from the book "Thinking, Fast and Slow: How to Avoid Irrational Decision-Making".

1. Noise and Bias

First of all, what are noise and bias?

Noise is the "random variability" of judgment: under different circumstances, people will draw different conclusions to the same problem.

On the one hand, different decision-makers have different judgments; on the other hand, even the same decision-maker may come to different judgments due to factors such as emotions and fatigue.

As for bias, it refers to decision-makers making judgments inappropriately based on past experience. Bias stems from people's overconfidence in their own experiences and viewpoints, which can lead to the "availability bias"-ignoring effective information that is not consistent with their own viewpoints.

Decision-making errors are the result of the combined effect of noise and bias: errors = noise + bias.

As shown in the figure below, using shooting as a metaphor, noise is the shooter's unstable hands that sometimes skew to one side and sometimes to the other; bias is that this gun has a problem and the bullets always skew in one direction. (Official WeChat platform ID: qlhclub)

The target in the upper left corner hits the bullseye with all five shots, and we say its overall noise and bias are small.

The five shots on the target in the upper right corner are everywhere, but overall and on average are evenly distributed around the bullseye, and we say its bias is small but the noise is large.

All five shots on the target in the lower left corner missed, but they missed in the same way, all clustered in the upper right corner of the target, and we say its bias is relatively large, but the noise is very small.

The five shots on the lower right target are both biased and scattered, and we say its bias and noise are large.

In more vivid terms: noise is the dispersion of judgment, which is inconsistent with each other; bias is systematic error, which skews in one direction.

2. The Impact of Noise

What is the impact of noise on decision-making? Perhaps beyond imagination.

Kahneman pointed out, "Noise is usually far beyond people's tolerance level, and they are completely unaware of it."

When people obtain investment information from news and commentary analysis, they are likely to be disturbed by surrounding noise and deviate from rational decision-making. The figure below shows the estimates of 50 accountants for payable taxes. Almost no one's analysis is similar. The noise is very large, and the smallest estimate differs from the largest estimate by nearly $12,000.

A study in the United States found that when predicting complex events such as social, political, and economic, judgments always outperform the difference between the "super forecaster" and the ordinary forecaster. 50% is due to the reduction of noise by the "super forecaster", 25% is due to the reduction of bias, and 25% is due to the increase in information. In other words, the reason why we are difficult to predict may be due to the failure to control noise, which is not as well-known as bias or information scarcity.

Therefore, it is likely that we fail to predict because of the control of noise, rather than the better-known biases or lack of information.

3. Noise in Investment

There are some "noise traders" in the financial market who package noise as effective information for trading. This interferes with investors' accurate understanding of the market, leading to chasing and killing, mispricing, and other situations.

Noise trading is naturally unacceptable, and we must avoid becoming "noise traders." But in the inevitable case of noise, we can also take measures to even use it for arbitrage.

Faced with noise, rational investors will collect and evaluate a large amount of information, understand the true value of securities, understand the trading behavior, bullish or bearish motives of other investors, form judgments on the short-term market trend, and thus see their own advantages.

For example, when the "noise trader" kills the market price, the rational investor will buy at a low price and wait for the market to approach the fair value; when chasing causes the price to be too high, it will resolutely sell and take profits. (Official WeChat account ID: qlhclub)

For example, induced by the market noise, we often think that finding cheap securities is extremely important, but we don't care about warehouse decisions that are rarely talked about. In order to achieve effective investment, in addition to discovering excess returns from cheap securities, we also need to determine warehouse positions, that is, how much capital to invest in this.

"Ignoring dosage, discussing toxicity is hooliganism", ignoring scale and discussing returns is also true. Many times, we spend a lot of time looking for mispriced symbols, but spend little time determining warehouse positions, which undoubtedly has a huge impact on returns.

4. How to Reduce Noise and Bias?

Noise and bias are so terrible, how can we avoid their adverse effects as much as possible?

(1) Pool wisdom

Integrating the wisdom of the group is the first step.

Everyone's judgment may have deviations in a random way, but when everyone's judgments are combined, the group error can be reduced and the decision accuracy can be improved.

Abraham de Moivre proposed the mathematical basis of this principle in the early 18th century: the standard error of the average value is equal to the standard deviation of the sample divided by the square root of the sample size. That is to say, the error decreases with the increase of the sample size.

If you want your investment decision to be as correct as possible, you can exchange ideas with several investment experts and refer to the average tendency of their opinions.

However, this method will achieve the best results only if the following three conditions are satisfied as much as possible:

1. Everyone has a relatively high level;

2. The style should not be too similar (eliminate style bias), otherwise the noise may be eliminated, but the bias may be amplified;

3. The most important point is that everyone does not communicate with each other and makes judgments independently.

Of course, you can choose not to accept other people's determinations, but you must realize that their determinations also affect the market.

(2) Use algorithms

An algorithm is a set of established procedures that helps you achieve your goals. A cake recipe is an algorithm. If you follow the recipe to mix the ingredients and bake the batter, you can make a delicious cake.

Some professionals do not recognize the effectiveness of algorithms. They believe that the causal relationship of algorithms is often too loose, and using algorithms is impractical. This is indeed reasonable, but the smartest approach is not to abandon algorithms completely, but to combine the standard actions of algorithms with their own initiatives. (Official WeChat account for Qilue Club: qlhclub)

Based on this, Kahneman proposes "disciplined intuition," which means using algorithmic system to systematically process problems, while using one's own judgment and intuition to assist in final decision making. There is evidence that such an approach that affirms algorithms and allows for subjective judgement can improve the overall quality of decision-making.

(3) Process-orientation: Adopting the "Mediation Assessment Procedure" (MAP)

The logic of MAP is to design more detailed second-level evaluation indexes based on the goals that the decision wants to achieve, and use these second-level indexes to better judge the rationality of the decision.

MAP has three steps:

1. Clarify what is important to determine the evaluation criteria;

2. Evaluate the likelihood of success for various types of decisions based on these criteria;

3. After completing all evaluations, discuss the final decision.

Compared with relying solely on intuition, MAP helps make wiser decisions. For investors in the public market, MAP can ensure consistency, rigor, and completeness in the investment process.

Editor/Jayden