Ben/what is probability by ben-herbst · Pull Request #141 · resampling-stats/resampling-with

ben-herbst · 2023-09-13T13:57:01Z

No description provided.

matthew-brett

Small comments and suggestions

matthew-brett · 2023-09-14T17:45:55Z

+In practice you will asked to provide answers based on data. For example, you may be given data about customer
+behaviour in a large bank and asked to develop a model that will provide the probability of default of the customers in
+the bank. This is an important problem for all financial institutions - if it does not have a good credit risk model,
+it will either loose money by being too conservative in the way it lends money, or loose money bay taking on too much


Suggested change

it will either loose money by being too conservative in the way it lends money, or loose money bay taking on too much

it will either loose money by being too conservative in the way it lends money, or loose money by taking on too much

matthew-brett · 2023-09-14T17:47:47Z

+behaviour in a large bank and asked to develop a model that will provide the probability of default of the customers in
+the bank. This is an important problem for all financial institutions - if it does not have a good credit risk model,
+it will either loose money by being too conservative in the way it lends money, or loose money bay taking on too much
+of a risk. You can opt for applying a sophisticated model such as a deep neural network but you are almost guaranteed


Won't they be confused by "deep neural network" here? Is there a more general way of saying this - such as:

Suggested change

of a risk. You can opt for applying a sophisticated model such as a deep neural network but you are almost guaranteed

of a risk. You can opt for applying an extremely complex "machine-learning" model with many parameters but you are almost guaranteed

matthew-brett · 2023-09-14T17:48:13Z

+the bank. This is an important problem for all financial institutions - if it does not have a good credit risk model,
+it will either loose money by being too conservative in the way it lends money, or loose money bay taking on too much
+of a risk. You can opt for applying a sophisticated model such as a deep neural network but you are almost guaranteed
+to come to grief. First study the problem and come to terms of all the many issues at stake. We speak of experience!


Better as?

Suggested change

to come to grief. First study the problem and come to terms of all the many issues at stake. We speak of experience!

to come to grief. First study the problem and come to terms of all the many issues at stake. We speak from experience!

matthew-brett · 2023-09-14T17:48:53Z

+Let's illustrate the idea with a simple example. The teacher asks little Annie to solve the following problem: Ten
+sheep are on this side of the road and one sheep crosses to the other side, how many sheep remain on this side? Annie
+knows the answer of course, and replies, correctly, none. This quite agitates the teacher and asks, there were ten sheep
+on this side of the road and one crosses over to the other side, how is it that you tell non remain? Annie replies,


Suggested change

on this side of the road and one crosses over to the other side, how is it that you tell non remain? Annie replies,

on this side of the road and one crosses over to the other side, why are you saying that none remain? Annie replies,

matthew-brett · 2023-09-14T17:49:31Z

+It is easy to get the arithmetic right, but as easy to get the problem wrong if you don't understand it.
+
+Please make sure you know what problem you have to solve. You may even run into situations where a company provides you
+with lots of data and then ask you to extract meaningful information from it. Our advice it, work with the company to


Suggested change

with lots of data and then ask you to extract meaningful information from it. Our advice it, work with the company to

with lots of data and then asks you to extract meaningful information from it. Our advice is, work with the company to

matthew-brett · 2023-09-14T17:54:16Z

+   These raise serious ethical questions that the practitioner should be aware of.
+
+   Returning to the criminal detection problem mentioned above, it failed. Let's think of what the model does. Since it
+   it given samples of photographs of criminals and non-criminals, i.e. each photograph comes with the label, `criminal`


Suggested change

it given samples of photographs of criminals and non-criminals, i.e. each photograph comes with the label, `criminal`

is given samples of photographs of criminals and non-criminals, i.e. each photograph comes with the label, `criminal`

matthew-brett · 2023-09-14T17:55:18Z

+## How is you model going to be used?
+
+The responsibility of the technical developer does not end with providing the model, or the analytics needed for the
+purpose. It is important to know how your model is going to be used. If you are to develop a system that need to


Suggested change

purpose. It is important to know how your model is going to be used. If you are to develop a system that need to

purpose. It is important to know how your model is going to be used. If you are to develop a system that needs to

matthew-brett · 2023-09-14T17:58:05Z

+is all about coding. Anyway, you will become so much more marketable if you learn the basics of solid software practices.
+We don't have the space to do it here but we do want to stress its importance. Always keep in mind the following:
+
+1. Use versioning control, we recommend using git. If you regularly push to the git repo this will protect you from


Suggested change

1. Use versioning control, we recommend using git. If you regularly push to the git repo this will protect you from

1. Use version control; we recommend using git. If you regularly push to the git repo this will protect you from

matthew-brett · 2023-09-14T17:58:23Z

+1. Use versioning control, we recommend using git. If you regularly push to the git repo this will protect you from
+   accidental software loss. It makes is also eay to share your code. You want other people to use your code, it make you
+   so much more useful!
+2. Ask someone elso to critically review your code. Even better if you work in an environment where there is a formal


Suggested change

2. Ask someone elso to critically review your code. Even better if you work in an environment where there is a formal

2. Ask someone else to critically review your code. Even better if you work in an environment where there is a formal

matthew-brett · 2023-09-14T17:58:51Z

+2. Ask someone elso to critically review your code. Even better if you work in an environment where there is a formal
+   system of code review.
+3. Read other people's code. You will learn a lot.
+4. Always tests for your code. This means that you run your code on small examples for which you know the answer. Every


Suggested change

4. Always tests for your code. This means that you run your code on small examples for which you know the answer. Every

4. Always add tests for your code. This means that you run your code on small examples for which you know the answer. Every

* Split out section on software practices to new chapter

ben-herbst added 2 commits September 12, 2023 16:38

* Added sections on understand your problem, etc

52ceea5

* Add a section on good software practices.

957804e

ben-herbst requested review from matthew-brett, pxr687 and stefanv September 13, 2023 13:57

matthew-brett reviewed Sep 14, 2023

View reviewed changes

ben-herbst added 2 commits September 15, 2023 09:58

* Incorporate pr suggestions

4cc6567

* Split out section on software practices to new chapter

- Add citations to criminal faces study

6f4c98a

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Ben/what is probability#141

Ben/what is probability#141
ben-herbst wants to merge 4 commits intoresampling-stats:mainfrom
ben-herbst:ben/What_is_probability

ben-herbst commented Sep 13, 2023

Uh oh!

matthew-brett left a comment

Uh oh!

matthew-brett Sep 14, 2023

Uh oh!

matthew-brett Sep 14, 2023

Uh oh!

matthew-brett Sep 14, 2023

Uh oh!

matthew-brett Sep 14, 2023

Uh oh!

matthew-brett Sep 14, 2023

Uh oh!

matthew-brett Sep 14, 2023

Uh oh!

matthew-brett Sep 14, 2023

Uh oh!

matthew-brett Sep 14, 2023

Uh oh!

matthew-brett Sep 14, 2023

Uh oh!

matthew-brett Sep 14, 2023

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

	it will either loose money by being too conservative in the way it lends money, or loose money bay taking on too much
	it will either loose money by being too conservative in the way it lends money, or loose money by taking on too much

	of a risk. You can opt for applying a sophisticated model such as a deep neural network but you are almost guaranteed
	of a risk. You can opt for applying an extremely complex "machine-learning" model with many parameters but you are almost guaranteed

	to come to grief. First study the problem and come to terms of all the many issues at stake. We speak of experience!
	to come to grief. First study the problem and come to terms of all the many issues at stake. We speak from experience!

	on this side of the road and one crosses over to the other side, how is it that you tell non remain? Annie replies,
	on this side of the road and one crosses over to the other side, why are you saying that none remain? Annie replies,

	with lots of data and then ask you to extract meaningful information from it. Our advice it, work with the company to
	with lots of data and then asks you to extract meaningful information from it. Our advice is, work with the company to

	it given samples of photographs of criminals and non-criminals, i.e. each photograph comes with the label, `criminal`
	is given samples of photographs of criminals and non-criminals, i.e. each photograph comes with the label, `criminal`

	purpose. It is important to know how your model is going to be used. If you are to develop a system that need to
	purpose. It is important to know how your model is going to be used. If you are to develop a system that needs to

	1. Use versioning control, we recommend using git. If you regularly push to the git repo this will protect you from
	1. Use version control; we recommend using git. If you regularly push to the git repo this will protect you from

	2. Ask someone elso to critically review your code. Even better if you work in an environment where there is a formal
	2. Ask someone else to critically review your code. Even better if you work in an environment where there is a formal

	4. Always tests for your code. This means that you run your code on small examples for which you know the answer. Every
	4. Always add tests for your code. This means that you run your code on small examples for which you know the answer. Every

Conversation

ben-herbst commented Sep 13, 2023

Uh oh!

matthew-brett left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants