Quantitative Neurobiology

Notes, assignments, and code for NEUROBIO 735 (Spring 2018).

Class details:

1/10 – 2/8:
Wednesday, Thursday
3:00 – 4:30
DIBS conference room

Exercises

Quizzes

GitHub

  1. In the generalized linear model, we use a linear combination \(\beta\) of the predictor variables \(X\) to model what?
    1. the mean of a distribution
    2. a function of the mean of the distribution
    3. the variance of a distribution
    4. the value of a new observation
    5. the observed data points
  2. Which of the following is a limitation of the GLM?
    1. It can only do linear regression.
    2. It assumes normally distributed errors.
    3. It cannot fit categorical or binary data.
    4. It cannot include categorical or binary predictors.
    5. It cannot directly model the variance of data.
  3. Which of the following is not an example of the split-apply-combine pattern?
    1. Calculating means of experimental and control groups.
    2. Normalizing each frequency band of a signal to have the same mean power.
    3. Blurring an image by applying a filter.
    4. Fitting a single GLM to data from multiple experimental groups.
    5. Removing artifacts from each channel of EEG data.
  4. For which of the following cases would you use logistic regression:
    1. Modeling white blood cell counts in an assay.
    2. Modeling the probability of a synapse releasing a vesicle.
    3. Modeling lever press rates for two groups of mice.
    4. Modeling response times in a contrast detection task.
    5. Modeling calcium fluorescence from a cell over time.
  5. Which of the following is required for split-apply-combine?
    1. An apply step that can be done in parallel.
    2. Categorical variables on which to split.
    3. An apply step with a single output.
    4. 1 and 2
    5. 2 and 3