# Copulas

refer to [An intuitive, visual guide to copulas](https://twiecki.io/blog/2018/05/03/copulas/)

```python
%matplotlib inline

import seaborn as sns
from scipy import stats
```

## Random Variable Transformation

```python
# uniform sampling
x = stats.uniform(0, 1).rvs(10000)
sns.distplot(x, kde=False, norm_hist=True)
```

![png](/files/-LpDhofC8Axcbx1sBsRb)

```python
# inverse CDF
norm = stats.norm()
x_trans = norm.ppf(x)
sns.distplot(x_trans)
```

![png](/files/-LpDhofE8BlosAX1u87i)

```python
h = sns.jointplot(x, x_trans, stat_func=None)
h.set_axis_labels('original', 'transformed', fontsize=16)
```

![png](/files/-LpDhofGr5rxO9Lo7HMT)

```python
# same work for beta
beta = stats.beta(a=10, b=3)
x_trans = beta.ppf(x)
h = sns.jointplot(x, x_trans, stat_func=None)
h.set_axis_labels('original', 'transformed', fontsize=16)
```

![png](/files/-LpDhofIaar6UA122vE8)

```python
# same work for Gumbel
gumbel = stats.gumbel_l()
x_trans = gumbel.ppf(x)
h = sns.jointplot(x, x_trans, stat_func=None)
h.set_axis_labels('original', 'transformed', fontsize=16)
```

![png](/files/-LpDhofKMtFlz0DnJ4K3)

```python
# convert an arbitrary distribution to the uniform (0, 1): CDF
x_trans_trans = gumbel.cdf(x_trans)
h = sns.jointplot(x_trans, x_trans_trans, stat_func=None)
h.set_axis_labels('original', 'transformed', fontsize=16)
```

![png](/files/-LpDhofMlzJWaz39xcwD)

## Adding correlation with Gaussian copulas

```python
mvnorm = stats.multivariate_normal(mean = [0, 0], cov = [[1., 0.5], [0.5, 1.]])
x = mvnorm.rvs(100000)
```

```python
h = sns.jointplot(x[:, 0], x[:, 1], kind = 'kde', stat_func = None)
h.set_axis_labels('x1', 'x2', fontsize=16)
```

![png](/files/-LpDhofOLiZ_8ipE_wsz)

```python
norm = stats.norm()
x_unif = norm.cdf(x)
h = sns.jointplot(x_unif[:, 0], x_unif[:, 1], kind='hex', stat_func=None)
h.set_axis_labels('Y1', 'Y2', fontsize=16)
```

![png](/files/-LpDhofQr1G5-ZE1_vo7)

```python
# transform the marginal
m1 = stats.gumbel_l()
m2 = stats.beta(a=10, b=2)

x1_trans = m1.ppf(x_unif[:, 0])
x2_trans = m2.ppf(x_unif[:, 1])

h = sns.jointplot(x1_trans, x2_trans, kind='kde', xlim=(-6, 2), ylim=(.6, 1.0), stat_func=None)
h.set_axis_labels('Maximum river level', 'Probability of flooding', fontsize=16)
```

![png](/files/-LpDhofS_zf13EQfqQID)

```python
# compare with the joint distribution without correlation
x1 = m1.rvs(10000)
x2 = m2.rvs(10000)

h = sns.jointplot(x1, x2, kind='kde', xlim=(-6, 2), ylim=(.6, 1.0), stat_func=None)
h.set_axis_labels('Maximum river level', 'Probability of flooding', fontsize=16)
```

![png](/files/-LpDhofU2dmjg8qTGDNG)

```python
```


---

# 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://mc.hohoweiya.xyz/genrv/an-intuitive-visual-guide-to-copulas.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.