STAT 451 Project Proposal

Group members: Anushka, Paige, Yumi, Kate

import opendatasets as od
import pandas as pd
import os
import numpy as np
import matplotlib.pyplot as plt
from sklearn import linear_model

1. Data descriptions

dataset description

“finch_beaks_1975.csv” and “finch_beaks_2012.csv”

• band is the identification tag for each bird
• species encoding as integers: 0 = scandens 1 = fortis
• Beak length, mm
• Beak depth, mm

“fortis_beak_depth_heredity.csv”

• Mid-offspr is the average beak depth between all of the offspring of a set of parents
• Male BD: the beak depth of male parent
• Female BD : the beak depth of the female parent

“scandens_beak_depth_heredity.csv”

• mid_parent: the average beak depth of mother and father
• mid_offspring: is the average beak depth between all of the offspring of a set of parents

2. Hypothesis

To connect these 4 datasets, we are interested in the evolution of the beak depth trait of the two different species of finches, scandens and fortis. fortis_beak_depth_heredity.csv contains hereditary information, which we can perform OLS regression to get a heredity index equivalent to the $R^2$ of the OLS regression model. We will do the same for the scandens_beak_depth_heredity.csv data to yield another heredity index for the scandens species. We hypothesize that for both species, the offspring beak depth will be different than their parents, and this will align with the heredity information.

3. Exploratory Analysis

beaks_1975 = pd.read_csv("data/finch_beaks_1975.csv", index_col=0)
beaks_2012 = pd.read_csv("data/finch_beaks_2012.csv", index_col=0)
fortis_heredity = pd.read_csv("data/fortis_beak_depth_heredity.csv")
scandens_heredity = pd.read_csv("data/scandens_beak_depth_heredity.csv")

beaks_1975['species']=(beaks_1975['species']=='fortis').to_numpy().astype(int)
beaks_2012['species']=(beaks_2012['species']=='fortis').to_numpy().astype(int)
beaks_1975 = beaks_1975.rename(columns={"Beak length, mm": "beak_length", "Beak depth, mm": "beak_depth"})
beaks_2012 = beaks_2012.rename(columns={"blength": "beak_length", "bdepth": "beak_depth"})

beaks_1975.head(10)

	species	beak_length	beak_depth
band
2	1	9.4	8.0
9	1	9.2	8.3
12	1	9.5	7.5
15	1	9.5	8.0
305	1	11.5	9.9
307	1	11.1	8.6
308	1	9.9	8.4
309	1	11.5	9.8
311	1	10.8	9.2
312	1	11.3	9.0

# compare the beak measurement in 2012 to 1975
beaks_1975.groupby('species')[['beak_length', 'beak_depth']].mean()

	beak_length	beak_depth
species
0	14.12092	8.960000
1	10.56519	9.171646

beaks_2012.groupby('species')[['beak_length', 'beak_depth']].mean()

	beak_length	beak_depth
species
0	13.421024	9.186220
1	10.517355	8.605372