Weather Comparisons: Seattle & New York City

Introduction

This report explores the temperature and precipitation data for Seattle and New York City. Using a Jupyter Notebook, you will explore and filter weather data to determine the temperature range at which both cities receive the most precipitation.

This report uses data collected by Weather Underground from July 1st, 2014 through June 30, 2015.

Data dictionary

Field	Description
date	The date of the weather record, formatted YYYY-M-D.
actual_mean_temp	The measured average temperature for that day.
actual_min_temp	The measured minimum temperature for that day.
actual_max_temp	The measured maximum temperature for that day.
average_min_temp	The average minimum temperature on that day since 1880.
average_max_temp	The average maximum temperature on that day since 1880.
record_min_temp	The lowest ever temperature on that day since 1880.
record_max_temp	The highest ever temperature on that day since 1880.
record_min_temp_year	The year that the lowest ever temperature occurred.
record_max_temp_year	The year that the highest ever temperature occurred.
actual_precipitation	The measured amount of rain or snow for that day.
average_precipitation	The average amount of rain or snow on that day since 1880.
record_precipitation	The highest amount of rain or snow on that day since 1880.
Month	The month extracted from date.
Year	The year extracted from date.

Terminology

Term	Description
Pandas DataFrame	An object of data, similar to a matrix. A Pandas Dataframe is a two-dimensional data structure of mutable size, arranged in rows and columns. Data types within columns can vary.
matrix	An array of arrays.
method chaining	A programming technique that allows the user to "chain" or attach subsequent methods to a single object.
data munging	The process of identifying and resolving invalid data in a dataset prior to analysis.

API reference

The following API reference section is divided by Python library.

pandas methods

Method	Description
.read_csv('foo.csv')	A method that reads a comma separated value (.csv) file(s) and saves it to a dataframe.
.head()	A method that displays the first five rows of a dataframe instance.
.tail()	A method that displays the last five rows of a dataframe instance.
.shape	A method whose output returns the number of rows and columns, respectively, in a given dataframe.
.columns	A method that returns the column names in a given dataframe.
.dtypes	A method that returns the data types of a given input.
.DataFrames	A method that converts its arguments to a DataFrame instance.
.merge()	A method that merges two dataframes on a specified column.
.isnull()	A method that detects missing values for an array-like object and returns an array of boolean values. Used to determine whether null values are present.
.sum()	A method that adds given inputs together.
.to_datetime()	A method that converts a given argument—such as a dataframe—to time.
dt.month	A method that extracts a given date's month as a number. For example, January=1.
dt.year	A method that extracts a given date's year.
set_index()	A method that takes in an argument—such as a dataframe—and lets you reset the index using a column value.

numpy methods

Method	Description
.nan	A method that provides placeholders for null values. Often used with .isnull() for data cleanup or data munging.

matplotlib.pyplot methods

Method	Description
.plot(figsize=(a,b))	A method for data visualization that returns a plot chart. The figsize parameter lets you control the size of the displayed chart.
.plot(kind='scatter', x='foo', y='bar', figsize=(a,b))	A method for data visualization that returns a plot chart. The kind parameter lets you define the kind or type of plot to render. The x parameter represents the value for the x-axis. The y parameter represents the value for the y-axis. The figsize parameter represents the size of the displayed chart.

For more information, visit the pandas documentation page.

Important Links

• Final Report Notebook
• EDA Notebook
• Sources: Weather Underground, 538 US Weather History

Note: This project uses CSV files referenced in the 538 US Weather History repo. The weather data for these files was collected by Weather Underground.