Mogic considers

Beyond the normalizing business matrix

There is a cause, there is a result, it is easy to understand.

Three causes, with shading effects on the results, are difficult to understand.

If there are more than ten causes, and if the first cause produces an intermediate result and then reacts in a chain reaction with other causes, it becomes very difficult to scrutinize.

The reason I made this a lengthy article is because I thought that the more factors there are, the more difficult it would be to identify the cause of the result.

However, multivariate analysis, which combines many factors in a model, sensitivity analysis, which simulates the blurring range by manipulating many factors this way and that, and others.

When we look at the business scene to apply these factors, there are still many, too many factors to be included.

I was doing some detailed research on multivariate analysis to see if I could improve the view even if there were many factors, and I found an application that I thought would be a good match for the logic of generative AI.

So, let's start with an overview of multivariate analysis

oooooooooooooooooooooooooooooooooooooooo
Multivariate Analysis
h ttps:// w.wiki/Dvsp

Multivariate analysis (multivariate analysis) is a method of summarizing the characteristics of multivariate data. Data summarization simplifies the characteristics of data and makes them easier to analyze.

Multiple Regression Analysis, Principal Component Analysis, Independent Component Analysis, Factor Analysis, Discriminant Analysis, Quantification Theory (Class I, II, III, IV), Cluster Analysis, Conjoint Analysis, Multidimensional Scaling (MDS).
oooooooooooooooooooooooooooooooooooooooooooooo

Then, as an example of application to reality, let's look at the methods of human geography.

oooooooooooooooooooooooooooooooooooooooooooooo
In human geography, multivariate analysis is an important technique in regional analysis.

Since the late 1950s, multivariate analysis has been used in the development of theories of econometrics.

In human geography, multiple regression analysis is used to determine connectivity among regions, principal component analysis is used to analyze the internal structure of cities, and factor ecological analysis and regional classification of isogenic and functional regions are conducted through factor and cluster analysis.

When performing multivariate analysis in a regional analysis, a geographic matrix is first created.

An attribute matrix would be created for setting up an isogenous region, and an interaction matrix for setting up a functional region, and then a multivariate analysis would be performed.

[Supplemental] Geographic Matrix https://w.wiki/DvvB
Berry (1964) takes attributes in the rows and regions in the columns, and the geographic event for the region in question is displayed at the intersection of each row and column (the matrix component). This allows all attributes for all regions to be displayed as a matrix.
oooooooooooooooooooooooooooooooooooooooooooooo

Geography matrices, or matrices of "attributes" and "regions" in geography, are used to prep the calculation.

The idea was that if we could construct a business matrix with components such as "job type" and "activity," then multivariate analysis via AI might yield useful insights.

What kind of future landscape can be seen beyond such thought experiments?

Yes, the first thing that comes to mind is the vast amount of business matrices that can be collected in real time globally, regardless of whether it is your company or another company.

Arithmetic operations using business matrices too grand for people to handle determine their position on the economic board and make them choose their actions.

Unfortunately, the logic is quickly propagated, imitated, and used by everyone in the blink of an eye.

If this is the case, the greatest risk is that they will converge as similar firms by blindly trusting in the same big and small logic.

In short, the selection of strategies may be the same as at the present time, regardless of whether large numbers of operations are performed or not, or, conversely, it may be that the goal is to achieve decisive excellence by finding very microscopic differences.