My biggest mistakes were never in my insights.
My mistakes were in overconfidence. ๐งต
#rstats #datascience #python
My mistakes were in overconfidence. ๐งต
#rstats #datascience #python
In business, I've made great regression models that have predicted how much sales we were going to make.
In fact, this helped me increase revenue from $3M to $15,000,000 per year at one of the companies I worked at.
In fact, this helped me increase revenue from $3M to $15,000,000 per year at one of the companies I worked at.
BUT my models were NOT perfect.
In fact, I'd argue that the BIGGEST flops were due to overconfidence.
Believing my model was better than it actually was.
Things that hurt me:
In fact, I'd argue that the BIGGEST flops were due to overconfidence.
Believing my model was better than it actually was.
Things that hurt me:
1. Anomalies (Outliers)
Not understanding what anomalies meant (and why they were important to the business problem)
Not understanding what anomalies meant (and why they were important to the business problem)
2. Skewed Data (Long tailed distributions)
This is risk.
This is risk.
3. Using error metrics based on in-sample data
In machine learning this is heresy...
But for some reason in time series it's commonplace.
Why???
In machine learning this is heresy...
But for some reason in time series it's commonplace.
Why???
This is where Bayesian can help.
Learning your models true confidence.
And, defining uncertainty in terms of business risk.
Learning your models true confidence.
And, defining uncertainty in terms of business risk.
I'd like to invite you to attend my FREE WORKSHOP on Bayesian, where I'm going to go over:
1. Why Bayesian is important to business
It's not accuracy that we need.
It's something even more important.
1. Why Bayesian is important to business
It's not accuracy that we need.
It's something even more important.
2. How to implement Bayesian for 90% of business problems
There's a specific pattern you can follow.
There's a specific pattern you can follow.
3. A case-study with code that shows you my strategies for analyzing Bayesian models
Talking about Bayesian is pointless.
We need to roll up our sleeves & apply Bayesian together.
Talking about Bayesian is pointless.
We need to roll up our sleeves & apply Bayesian together.
Loading suggestions...