Hypothesis

A hypothesis is a pattern or rule about a process in the world that can be tested. We use hypothesis testing to determine if a change we made had a meaningful impact or not.
You can use hypothesis testing to help you determine:
  • if a new banner ad on a website caused a meaningful drop in the user engagement,
  • if raising the price of a product caused a meaningful drop in sales,
  • if a new weight loss pill helped people lose more weight.
Observing a decrease in user engagement or sales after instituting a change doesn't automatically imply that the change was the cause. Hypothesis testing allows us to calculate the probability that random chance was actually responsible for the difference in outcome. Every process has some inherent amount of randomness that we can't measure and understanding the role of chance helps us reach a conclusion that's more likely to be correct.
We first set up a null hypothesis that describes the status quo. We then state an alternative hypothesis, which we used to compare with the null hypothesis to decide which describes the data better. In the end, we either need to:
  • reject the null hypothesis
  • fail to reject the null hypothesis
We can frame each of the studies above as these rival pairs of hypotheses:
  • if a new banner ad on a website caused a meaningful drop in the user engagement:
    • null hypothesis: users who were exposed to the banner ad spent the same amount of time on the website than those who weren't.
    • alternative hypothesis: users who were exposed to the banner ad spent less time on the website than those who weren't.
  • if raising the price of a product caused a meaningful drop in sales:
    • null hypothesis: the number of purchases of the product was the same at the lower price than it was at the higher price.
    • alternative hypothesis: the number of purchases of the product was lower at the higher price than it was at the lower price.
  • if a new weight loss pill helped people lose more weight:
    • null hypothesis: patients who went on the weight loss pill lost no more weight than those who didn't.
    • alternative hypothesis: patients who went on the weight loss pill lost more weight than those who didn't.

Comments

Post a Comment

Popular posts from this blog

Support Vector Machines

Image
Support Vector Machines A Support Vector Machine (SVM) is a powerful supervised machine learning model used for classification. An SVM makes classifications by defining a decision boundary and then seeing what side of the boundary an unclassified point falls on. Decision boundaries get defined,  by using a training set of classified points.     Decision boundaries exist even when your data has more than two features. If there are three features, the decision boundary is now a plane rather than a line.   As the number of dimensions grows past 3, it becomes very difficult to visualize these points in space. Nonetheless, SVMs can still find a decision boundary. However, rather than being a separating line, or a separating plane, the decision boundary is called a separating hyperplane . Optimal Decision Boundaries In general, we want our decision boundary to be as far away from training points as possible. Maximizing the dis...
Read more

ACCURACY, RECALL, PRECISION, F1 SCORE

ACCURACY, RECALL, PRECISION, F1 SCORE Classifying a single point can result in a true positive ( truth = 1 , guess = 1 ), a true negative ( truth = 0 , guess = 0 ), a false positive ( truth = 0 , guess = 1 ), or a false negative ( truth = 1 , guess = 0 ). Accuracy measures how many classifications your algorithm got correct out of every classification it made. Recall measures the percentage of the relevant items your classifier was able to successfully find. Precision measures the percentage of items your classifier found that were actually relevant. Precision and recall are tied to each other. As one goes up, the other will go down. F1 score is a combination of precision and recall. F1 score will be low if either precision or recall is low. The decision to use precision, recall, or F1 score ultimately comes down to the context of your classification. Maybe you don’t care if your classifier has a lot of false positives. If that’s the case, precision doesn’t matter as m...
Read more

Random Forest

RANDOM FOREST A random forest is an ensemble machine learning model. It makes a classification by aggregating the classifications of many decision trees. Random forests are used to avoid overfitting. By aggregating the classification of multiple trees, having overfitted trees in a random forest is less impactful. Every decision tree in a random forest is created by using a different subset of data points from the training set. Those data points are chosen at random with replacement , which means a single data point can be chosen more than once. This process is known as bagging . When creating a tree in a random forest, a randomly selected subset of features are considered as candidates for the best splitting feature. If your dataset has n features, it is common practice to randomly select the square root of n features. Boosting Steps : Draw a random subset of training samples d1 without replacement from the training set D to train a weak learner C1 ...
Read more