Tackling Extreme Values: A Step-by-Step Pearson Type III Guide

Ready to level up your data game?

Discover how the Pearson Type III Outlier Detection Method makes spotting unusual values in your data easy and accurate!

This tutorial breaks down the method with clear steps, visuals, and real-world insights—perfect for students, researchers, and data enthusiasts.

Don’t miss out—click play and start mastering outlier detection today!

Just hit subscribe on my YouTube channel: https://www.youtube.com/@MrinmoyMajumderor
You can also get updates straight to your inbox—sign up for my newsletter here: https://veryshorttermcourse.substack.com/
Looking for more tutorials like this? Check out the Very Short-Term Course on Hydroinformatics: http://www.baipatra.ws

Free Tutorial on Log Pearson Type III Outlier Detection Method

The Log-Pearson Type III Outlier Detection Method is a key tool in hydrology used to identify unusually high or low flood events that deviate significantly from the normal flow record. Based on Bulletin 17B guidelines, this statistical method evaluates annual peak discharge data using logarithmic transformations and parameters such as the mean, standard deviation, and skew coefficient of the data series.​

By detecting and appropriately treating outliers, analysts can ensure that flood frequency analyses remain accurate and not distorted by extreme or erroneous values. The procedure computes critical thresholds for high and low outliers, allowing hydrologists to differentiate between true extreme events and statistical anomalies in a river basin dataset.​

If you’re interested in understanding how this method works in practical flood analysis, watch the detailed explainer video here: Introduction to Log Pearson Type III Outlier Detection Method.

My Personal Page
Author,Lecture Notes on MCDM(Indian Link)
Lecture Notes on MCDM(Global Link)
My Kudos Profile

Ecourse on Multi Criteria Decision Making Methods

1) ELECTRE : ELIMINATION AND CHOICE TRANSLATING REALITY

2)FMAE : FAILURE MODE EFFECTS ANALYSIS

3)MAUT : MULTI ATTRIBUTE UTILITY THEORY

4&5)PROMETHEE : PREFERENCE RANKING ORGANIZATION METHOD FOR ENRICHMENT EVALUATION (One and Two)

6)RA : RELIABILITY ANALYSIS

7)WSM : WEIGHTED SUM METHOD

8)WPM : WEIGHTED PRODUCT METHOD

9)DELPHI Method

All these are free video tutorials. For case studies and project ideas please upgrade to Paid Member.

Click here to procure in INR:

My Personal Page
Author,Lecture Notes on MCDM(Indian Link)
Lecture Notes on MCDM(Global Link)
My Kudos Profile

Non Linear Regression Techniques that You Must Know As a Researcher of Science and Engineering Streams

Notes on

Dynamic Programming, Recursive Equation, Floyd Warsaw Algorithm, and KKT Conditions

Click here to download the tutorials

Author, Lecture Notes on MCDM(Indian Link)
Lecture Notes on MCDM(Global Link)
My Kudos Profile

EIS introduced It's PPTSHARE Service

The main objective of this PPTSHARE service is to get free but targeted views for the work that you presented in a conference or webinar or in invitational lectures.

My Personal Page
Author,Lecture Notes on MCDM(Indian Link)
Lecture Notes on MCDM(Global Link)
My Kudos Profile

Seven Most Popular Techniques That is Used to Analyze Research Output

In any research work, the basic steps are more or less the same. In their blog ScienceBuddies has correctly summarized the basic steps of research, as the following :

1. Ask a Question

2. Do Background Research

3. Construct a Hypothesis

4. Test Your Hypothesis by Doing an Experiment

5. Analyze Your Data and Draw a Conclusion

6. Communicate Your Results

However, the techniques you will employ at each step may differ. The efficiency of the techniques you use to achieve your desired goal will determine the optimality and rightness of your research output. If you carefully examine the six steps outlined above, you will quickly realize that the most important step is the fifth, which allows you to draw a conclusion with all of the necessary support and documentation retrieved from the various techniques available to you. The quality of your research output will be determined by the techniques you employ in this step.

This blog post tries to collect the most used techniques of research that you can use to analyze your results. The techniques are as given below :

1)Nonlinear time series models(NLTM),

2)State-space models(SSM)

3)Bayesian methods(BM)

4)Extreme value distributions(EVD)

5)Hierarchical models(HM)

6)Nonlinear regression(NLR), and 

7)MCMC or Markov Chain Monte Carlo Method. 

Click on the techniques to learn more about them and how they are used to identify and explain the most significant findings in any research study (I tried to collect a link that would give you a basic understanding of the technique). If you want to learn the techniques in greater depth, it is best to consult books/tutorials on the subject.)

The graph below shows how many publications—measured in lakhs—used the techniques and had their work accepted by reputable journals published by Elsevier in 2022.

The graph unambiguously illustrates the popularity of the methods used to publish papers in Elsevier journals 3,01,002 times in just one year. State Space Models, or SSM, were found to be the technique most frequently used among the seven techniques (equal to 1,28,556).

As a result, you can organize your research well in advance and may utilize SSM to evaluate the results of your work in preparation for publication in reputable journals.

That's all for today,

See you in the next post.

You may comment on the method you use to analyze your research outputs.

Subscribe to Research Geek (part of hydrogeek) where info like the above is shared.

My Personal Page
Author,Lecture Notes on MCDM(Indian Link)
Lecture Notes on MCDM(Global Link)
My Kudos Profile

Seven Research Area where MCDMs are used for better results

The use of MCDMs, or Multi-Criteria Decision Making Methods, in various aspects of research has increased dramatically in recent years. The total number of papers on MCDMs is 1298 in 2022, compared to only 51 in 1999. That translates to a 25.45-fold increase in the number of research papers published in journals of Elsevier Publishers.

If we categorize the papers based on the objective of MCDM application, we can broadly divide the published papers into the seven perspectives listed below. MCDMs are used for other purposes as well, but the number of papers on those topics is limited.

Location/Site Selection

MCDMs are frequently used to identify an ideal site or location for the installation or relocation of some industrial units, power plants, business centers, market complexes, and so on. The main objective of MCDM applications remains to consider some factors and identify the best location among the multiple locations available for selection.(Example : Nagababbu et.al.,Bairagi, B)

Performance Evaluation

MCDM techniques were used in several papers to analyze the performance of a product, an industry, or a proposed project. In a few cases, the application of some techniques was used to evaluate some models/formulae/applications.(Example: Nkuna et.al., Stevic et.al.)

Benchmark Development

The creation of benchmarks is critical in performance analysis. The goal included developing benchmarks for evaluation using MCDM techniques, and then applying the technique to evaluate the performance of some products/processes/etc. by comparing the performance to the developed benchmark. (Example: Lohakare et.al., Sumrit et.al.)

Development of Decision Support Systems

MCDMs are also used in the development of integrated decision support systems. This technique is primarily used for the development of the DSS's logic, which is used by the system to make decisions. DSS are used to achieve a variety of goals such as supplier selection, material selection, disaster management, and preventive measurements. (Ghazi et.al., Namin et.al.)

Vulnerability/Risk Analysis

Another objective where MCDMs are used alternatively is vulnerability/risk analysis. Vulnerability is the process of identifying and analyzing potential issues that could negatively impact the system's key objectives. Risk is the process of identifying and analyzing potential issues that could negatively impact the system's key objectives. Both vulnerability and risk involve a lot of decision-making through prioritization, comparing with benchmarks, or feature selection. MCDM is a technique for decision-making objectives such as prioritization, feature selection, and so on. That is the reason why there is a large number of papers on the application of MCDMs in vulnerability or risk analysis. (Mondal et.al., Tanim et.al.)

Index Development

An index is a metric or indicator about something. In the stock exchange, an index is used to measure the performance of any particular aspect of the market. The index developed for ecological studies is used to quantify and summarise ecological conditions at multiple levels. The weights of significance assigned to the input parameters based on their impact on the output are used to create an index. The primary goals of all compensatory MCDMs(Compensatory and Outranking techniques comprise the two categories of MCDMs. The distinction between the two is explained in my MCDM lecture notes.) (such as AHP, ANP, MACBETH, and so on) are to determine the weights of the input parameters so that the output can be used as a single point or number indicator of the inputs. As a result, MCDMs are widely used in the development of indicators. (Example: Gavalas et.al., Pathan et.al.)

Prioritization

The process of determining the level of importance and urgency of a task, system, event, etc. is known as prioritization. Prioritization is broadly equivalent to a process for identifying the best solution among many available solutions. The MCDM techniques were discovered to be ideal for use in prioritizing a list of solutions or options in order to identify the most optimal among them. (Example: Sarkar et.al., Ghasemlouniaa and Utlub)

Examining the published papers revealed that many times, multiple MCDMs were used in a hybrid mode by researchers rather than just one. GIS is sometimes used to visualize the technique's output. The most commonly used technique under MCDM is AHP, which is followed by Fuzzy Logic. Even in situations where there is a high degree of certainty, different MCDMs can produce inconsistent rankings of the alternatives. When two different MCDM techniques are used on the same decision problem, they can produce two different solutions. Because of this, the two most difficult tasks for a new researcher when developing an MCDM-based methodology are the selection of an appropriate analytic method and the interpretation of the MCDM results.

Thank you for reading the article. If you have any comments kindly hit them to me.

@Mrinmoy

Subscribe

Author, Lecture Notes on MCDM(Indian Link)

Lecture Notes on MCDM(Global Link)

My Kudos Profile