Subjective Probability: Function, Applications & Examples

Introduction

Subjective probability is a type of probability wherein a specific outcome is likely to happen based on your judgment or experience. It helps you predict the outcome of an event either by referencing things that you have learned so far or based on your own experience.

It is the exact opposite of objective probability, which measures the history of gathered data or recorded observations to predict if an event is going to occur or not. Subjective probability does not make use of mathematical calculation or data analysis; it rather depends upon your gut feeling to predict the outcome.

Subjective probability outcome will never be the same for two people as each person may have a different opinion or thinking about a particular event. For example, if two people were asked to predict how you would react in a specific situation, you will certainly hear different answers.

Since the person closer to you knows you better, he/she would give a response based on your nature. The other person may consider different factors while predicting your response.  

How Does the Subjective Probability Work?

Subjective probability has a high degree of personal bias, and the outcome of the probability is different for each person. In general, the probability is obtained by collecting the quantitative information or data and then interpreting the output by using some mathematical calculation or formula, which usually gives you a more accurate answer.

For example, you can predict the output of flipping a coin. There are always 50%-50% chances that the coin will land with a head and tail up.

On the other hand, the subjective probability is highly flexible and may not give you the correct answer as it is highly dependent on personal judgments held by each person. Judgments are based on an individual’s own experience and views. It often differs from person to person as they are subjective and based on how they interpret the situation given to him/her.

Though it does not use any statistical analysis or mathematical calculation, you can illustrate subjective probability as follows:

Probability (x) = degree of personal belief that x is true

Where x is an event, situation, or condition

Read: Logistic Regression in R

Uses of Subjective Probability

You might be wondering where we use subjective probability when it is so flexible and has no logical reasoning. Well, some industries find it useful and use these predictions to drive their business goals.

The subjective probability approach is used in multiple industries for decision-taking, such as marketing, economics, business, etc. For example, a sales manager predicts that there are 70% chances of getting the order for which his/her company has bid. Repeated tests or calculations cannot evaluate this percentage.

Some real-life examples where we use subjective probability are:

• Job interviews outcome
• Employee promotion
• Performance incentives
• Business sale
• Disadvantages of Subjective Probability

Subjective probability is highly affected by an individual belief or judgment about the likelihood of an event. The following are the disadvantages of subjective probability:

• The predictions are not backed up by logical reasoning or statistical calculation; it is always based on a high degree of bias.
• Subjective probability fails to meet the complex calculations.
• The outcome is never the same for an event or situation. For example, two or more individuals given the same situation may arrive at different outcomes, i.e., there may be different factors considered by the individuals for the same event.
• It must follow a few conditions to be workable. For example, when you predict the percentage of an event, whether it will occur or not, it must sum up to 100%.

Must Read: Types of Regression Models

Examples of Subjective Probability

The following examples clearly state how subjective probability outcome differs for each person.

Example 1

You are a huge fan of Virat Kohli, and the world cup cricket series is about to begin in a few days. You have been asked to predict the chances of India winning the World cup series. While there is no mathematical calculation or data to back up your predictions, you will still vouch for your favourite player or team in the actual percentage. For example, there is a 90% chance that India will win the World cup series.

Example 2

You have been asked to predict the outcome of a flipping coin, whether it will land with head or tail up. Though, the mathematical calculation says that there is a 50% chance that it will land head up and a 50% chance that it will land tails up. In subjective probability, the percentage of your prediction may change based on the previous flips.

If the same coin has been flipped 15 times in the past and has given ten times heads up and five times tails up. You will say that there is a 75% chance of landing heads up. Though it is mathematically incorrect, your experience has created a situation that compels you to predict using subjective probability.

Example 3

The weather department is predicting that it will rain in the next 2 hours based on wind pattern, weather situations, and their software analysis. But you may have the same predictions of rain in the next 2 hours based on your experience with weather or rain.

Example 4

You have fallen sick and want to visit your family doctor. You want to predict how much money you should take while visiting a doctor. The last time when you visited him, the doctor was charging 500 Rs as a consultation fee.

But one of your family members informs you that he has upgraded his office to add facility, due to which his consultation charges might have gone up. You are now left with two options; either go by your budget or go without the idea of what the cost is going to be and end up spending more money. You will go by your gut instinct and choose the first option.

Conclusion

To sum it all up, subjective probability is a type of probability based on individual knowledge, understanding, and experience of the likelihood of an event.  These predictions might be true if they are biased free and come up with some logical reasoning. But, there are situations, as explained throughout this article, that demands judgments or experience rather than calculations.  

If you’re interested to learn more about machine learning, check out IIIT-B & upGrad’s PG Diploma in Machine Learning & AI which is designed for working professionals and offers 450+ hours of rigorous training, 30+ case studies & assignments, IIIT-B Alumni status, 5+ practical hands-on capstone projects & job assistance with top firms.