# SVD

## Introduction

Calculates coordinations by reducing dimensionality using SVD (Singular Value Decomposition).

## How to Access?

You can access from 'Add' (Plus) button.

![](/files/-M4oNDDGM_aH1cIlex8d)

## How to Use?

#### Column Selection

![](/files/-M4oNDDIyZnxsoOh5aVO)

* Subject Column - Set a column to be vectorized by reduced dimensions.
* Key Column - Set a column to use as dimensions.
* Value Column (Optional) - Set a column to use as values. If empty, count of subject and key pair is used.
* Fill (Optional) - The default is 0. This is what should be used for missing value in groups.
* Aggregate Function (Optional) - The default is mean. This is how duplicated data should be aggregated.

### Parameters

* Output Type (Optional) - The default is "group".
  * "group" is to see how groups are projected to a new coordinations. This corresponds to u matrix of svd.
  * "dimension" is to see what direction the new coordinations face. This corresponds to v matrix of svd.
  * "variance" is to see how data is scattered in each of the new coordinations. This corresponds to d vector of svd. itself.
* Number of Dimensions (Optional) - The default is 3. Set number of dimensions to reduce to. This must be less than the number of subjects and the number of keys.
* Centralize Data - The default is TRUE. Set if data should be centralized to the origin.
* Output Format (Optional) - The defualt is "wide".
* "wide" - The components will be in columns.
* "long" - The component will be in row.

Take a look at the [reference document](https://stat.ethz.ch/R-manual/R-devel/library/base/html/svd.html) for the 'svd' function from base R for more details on the parameters.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.exploratory.io/statistics/svd.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.