There are several statistical methods that can be used to detect a relationship between two metrics. Some commonly used methods include: 

Pearson's correlation coefficient: This measures the strength and direction of the linear relationship between two variables. It ranges from -1 to 1, where -1 indicates a perfect negative linear relationship, 0 indicates no relationship, and 1 indicates a perfect positive linear relationship.

Spearman's rank correlation coefficient: This is similar to Pearson's coefficient, but it is used when the two variables are not necessarily linearly related. It also ranges from -1 to 1, with the same interpretation as Pearson's coefficient.

Chi-squared test: This is a statistical test that is used to determine whether there is a relationship between two categorical variables. It calculates the difference between the observed frequencies of the variables and the expected frequencies if there was no relationship, and determines whether the difference is statistically significant.

T-test: This is a statistical test that is used to determine whether there is a significant difference between the means of two groups. It can be used to compare two metrics if the data can be divided into two groups based on some criteria.

In general, the appropriate method to use depends on the type of data and the nature of the relationship between the two metrics. It's important to carefully consider the data and the research question at hand before choosing a statistical method.

 A weighted average is an average that takes into account the relative importance or weight of each value in a data set. A simple average, also known as the arithmetic mean, is an average that treats all values in the data set as having equal importance.

Here is an example to illustrate the difference between the two. Let's say you have a data set with the following values: 2, 4, 6, and 8. The simple average of this data set would be (2+4+6+8)/4 = 5. The weighted average, on the other hand, would be calculated by multiplying each value by its corresponding weight and then dividing by the sum of the weights. For example, if the weights were 0.5, 1, 1.5, and 2, the weighted average would be (20.5+41+61.5+82)/(0.5+1+1.5+2) = 6.

As you can see, the weighted average takes into account the relative importance of each value in the data set, while the simple average does not. In this case, the weighted average is higher than the simple average because the higher values in the data set have more weight.

 It really depends on the data set and what you are trying to do with it. Here is a brief explanation of each of these statistical measures:

Mean: The mean is the average of a set of numbers. It is calculated by adding up all the numbers in a data set and dividing by the number of items in the set.
Median: The median is the middle value in a data set. To find the median, you need to first order the data set from least to greatest, and then find the value that is in the middle of the set. If there is an even number of items in the data set, the median is the average of the two middle numbers.
Mode: The mode is the most common value in a data set. It is the value that appears most often in the data set.

In general, the mean is a good measure of central tendency when the data set is continuous and symmetrical. The median is a good measure when the data set is skewed or has outliers. And the mode is a good measure when the data set is categorical.

Here are some examples of scenarios where one of these measures might be better suited than the others:
If you are analyzing the heights of a group of people, the mean would be a good measure to use because height is a continuous variable and it is unlikely that there are any extreme outliers in the data set.
If you are analyzing the ages of a group of people, the median would be a better measure to use because age is a continuous variable and it is likely that there are some extreme outliers in the data set (e.g. very young or very old individuals).
If you are analyzing the favorite colors of a group of people, the mode would be the best measure to use because the color is a categorical variable and there is no inherent order to the values.

It's important to note that these are just general guidelines, and the best measure to use really depends on the specific data set and what you are trying to do with it. In some cases, it might be appropriate to use multiple measures to get a more complete picture of the data.

10 Benefits of AI in the Banking and Financial Sector

Artificial Intelligence has almost become a disrupter in every industry and so banking is not the exception to that. The banking sector became more customer-centric and technologically powerful when banks started using Artificial Intelligence in their web, applications, and online services.
AI in banking helps banks to make decisions based on the data and reduces the cost by increasing productivity. According to a report of Business Insider, approximately 80% of banks are aware of the potential benefits of Artificial Intelligence in banking. And another report says that banks are projected to save around $450 Billion by implementing AI in banking.

Uses Of AI in Banking and Finance
AI technologies have become the most important part of any industry in today’s world. Here are some AI applications in the banking industry.

Cybersecurity – Every day a huge number of digital transactions take place as people use to pay money, deposit money or checks, withdraw money. And in today’s world people uses online banking applications or third-party applications for doing all that. So there is a huge chance of cybersecurity. So AI in banking is an increasing need for cybersecurity and fraud detection AI can help banks to track the loophole in banking and minimize the risk.
Chatbots – As people don’t want to visit banks nowadays so whenever they face some issues and have some queries then they want some assistance from the bank side. And here chatbot comes to the picture. By integrating chatbots into banking apps, the banks can ensure that they are available for the customers 24*7.
Loan and Credit Decision – An AI-based loan and credit system can check the behavior and pattern of a customer with limited credit history to determine if it is worthy for the bank to give credit to them or not. AI in banking helps the bank to make more profitable, informed, and safe loan and credit decisions. No one can deny that credit report systems are often riddled with errors, but when AI came to the picture, it changed the game.
Tracking Market Trends – AI in banking sectors helps banks to process a huge volume of data and predict the latest market trends. It helps to evaluate the market sentiments and suggest the best investment options in stocks, bonds, and cryptocurrency.
Data collection and Analysis – Bank records millions of transactions every day which generates a huge volume of information. So, structuring and analysis of these huge amounts of data are impossible manually. AI in the banking sector helps to structure and analyze these data which leads to correct decision making.
Customer Experience and Customer Acquisition – Customers are always seeking a better experience, which leads to customer acquisition. For example, ATMs help customers to deposit or withdraw money 24*7. So, customers don’t need to go to the bank. Nowadays customer completes their KYC and can even open their bank account from home. So, they don’t need to go to the bank for that. AI in banking helps to capture this customer information accurately.
Risk Management - Global factors like currency rate fluctuations, political critical situations, or natural disasters have a serious impact on the banking industry. It's most important to take decisions at that time. So, AI in banking helps to analyze the risk of that situation and also provides a solution for that.  
Regulatory Compliance – Where it’s about money then there comes regulation. The government uses regulatory authority. Bank maintain their internal regulatory authority. AI in banking with the help of NLP helps the bank to improve its decision-making process. Which makes their operation faster and more efficient.
Predictive Analytics – Two major parts of AI general-purpose semantic and natural language applications are broadly used for Predictive Analytics. Artificial Intelligence can easily identify any specific pattern in data and correlation between the data. Which leads to some unknown sales opportunities. These metrics can even help to cross-sell any banking product like a credit card.  
Process Automation – RPA or Robotic Process Automation algorithm in Artificial Intelligence increases operational accuracy as well as efficiency and reduces costs by automating time-consuming repetitive tasks. Some banks are presently using Robotic Process Automation to increase efficiency and boost transaction speed.

BLOCKCHAIN FUNDAMENTALS...

Even if you are using the Old keypad phones, I am sure you have heard about

Blockchain,

Web 3.0,

NFTs,

Bitcoins,

DAOs.

So many new terms. It's time to understand them all.

What is Blockchain?

Blockchain is a database that is Decentralized and Digitally Distributed and not managed by a single company. By making a peer-to-peer database Blockchain is managed by multiple people.

Why do we even need Blockchain, which is that complicated?

In one word, to build the faith and to reduce the compliance charge. Blockchain allows digital transactions and records digital information. As this technology is not managed by any company so nobody can edit any information which itself builds trust in the blockchain.

let's understand this with an example -

Imagine you are traveling around the world and at some point, you are out of money. You ask your dad to send you some money. Let's assume 15000 bucks.

Now your dad instantly checks out his bank account and transfer the money. Then he texted you that he wired you the money and you will get it soon.

Now let's introspect what happened behind the scene...

Your dad send you the money and you got it, but the bank worked as a mediator. The moment your dad sends the money the bank registered the transaction, both the account number and date and time of the transaction. Writing the whole transaction from person X to person Y with a date and time has a cost. That is the bank's infrastructure cost. Because anytime anybody wire money to someone then the bank has to pay their server powers, human powers, etc. Which bank takes as transfer charge from the sender. In India, digital payment still is not that much penetrated. So to encourage digital payments, banks still charge noting from the sender. Whenever a huge number of people will start using digital payment then definitely bank will charge some percentage. Which they already do in developed countries.

Now understand the problem-

If your dad gives you 15000 bucks hand to hand then it's cost nothing. Then why can't we do the same online? Why can't we send it directly to anyone?

Because most of the time we send money to random people. We use online payments while buying online products too. So we want to keep a record. Here come TRUST ISSUES.

By keeping the record, banks and other mediators build that trust for each and every transaction we made.

But can we trust them fully? As all the banks and other mediators are owned by someone so can't trust them fully. Our transaction is not private anymore. Anybody can check it. So here comes the BLOCKCHAIN.

Initially, blockchain was made to serve as a mediator for every transaction we made to build trust. All the cryptocurrencies are made on this blockchain technology. Nowadays even many banks are adopting this technology.

And how does Blockchain do this?

For every transaction, blockchain makes a "block". Which generates a unique identity. Nobody can track that and by adding all the blocks blockchain makes a "chain".

There is a lot more than wiring money we can do with blockchain technology.

• personal identity security
• personal data security
• NFT marketplace
• Supply chain logistic monitoring

So, as I said earlier, every company has a server or multiple servers. So they keep the data within their servers and they can easily manipulate that. But blockchain runs at multiple PCs of multiple people spread across the world.

Now the next masterclass question is how do we know every transaction is real?

By checking the record of every block of a chain that is processed and verified.

But how can we still trust it? Who runs the verification?

That is the next topic. Will update it in the next 3 days. Follow me to know more about technology, education, and personal finance.

for any work or query mail me on - sourav.nag094@gmail.com