Important Information about the Final Exam
Important Information about the
Final Exam
Copyright By PowCoder代写 加微信 powcoder
Final Exam
2.5 hours exam:
Plus 0.5 hour to solve any tech issues
This means you will all have 3 hours to complete the exam
Combination of multiple choices, true/false, short answers, code
challenges
All answers must be submitted through ED
No immediate feedback (MCQ/true false) or test cases (code challenges)
You can use PyCharm if you want
Just make sure you submit your codes through ED
Make sure you go through the practice exam before the actual final
Material covered in the Final Exam
The final exam will cover:
Lectures 1 to 9
Any concept discussed in the quizzes/code challenges and projects
All modules under lectures (i.e., lec_XXX.py)
Simplified version of the event study:
More information on the next slides. . .
Lectures 10 and 11 will not be included
There is no need to go over the modules in the event_study package
Material covered during Week 9
Everything discussed in class during Week 9 is fair game:
Including both week9_slides_XXX.py modules
Make sure you go over the solutions in Dropbox
Including both week9_slides_XXX.py modules
Solutions will be posted on Thursday (Week 10)
Material covered during Week 10
Simplified event study: Everything up to (and including) Step 3
Class slides (the PDF in ED called “Week 10”):
All slides up to and including “Interlude: Groups and selection using
You can skip the remaining slides (starting with “Step 4: Calculate CARs
Modules under webinars:
Everything in module week9_slides_XXX.py (we finished this module in
From the module week10_slides_main.py:
Only the functions step2 and step3
These functions summarise many concepts discussed during weeks 7 to 10
Comments about individual lectures
Below, I emphasise a few important topics in each lecture
Unless explicitly stated, all material is fair game
Lecture 1: Financial analysis with Python: Downloading
stock prices
Motivating lesson
You won’t need to download any data during the final
The yfinance module will not be part of the exam
Lecture 2: Python: The building blocks
Important concepts:
Classes, instances, methods
Assignment statements
Names: both rules and conventions
E.g., variable names either all lowercase or all uppercase (constants)
No need to memorise reserved keywords or builtin function names
Assignment statements and copies
Lecture 3: Python: Control flow
Important concepts:
Loops, conditionals, functions, comprehensions
Loops (both for and while)
How to loop over data structures
Comprehensions
Difference between is and ==
Topics not covered:
Generator expressions (at the end of Lec 3.4)
Lecture 4: Working with modules
Important concepts:
Modules and packages
Accessing other modules
Namespaces
How to open a file using a context manager
Topics not covered:
Docstrings and styles (Lect 4.5)
But you need to know how to read a docstring
Lecture 5: Introducing Pandas (I)
Important concepts:
Series, indexes, data frames
How to construct series/data frames (nothing fancy. . . )
df = pd.DataFrame(data, index, columns)
ser = pd.Series(data, index)
Sorting (both by index and values)
Modifying objects in place (inplace=True)
Make sure you know
What these attributes represent: df.index, df.columns, ser.index
What df.info() does
Lecture 5: Introducing Pandas (II)
From Lecture 5.3 (Pandas and Numpy), you only need to know:
What numpy.nan means (i.e., not-a-number, missing)
The following data types (dtype):
Pandas Python Notes
int64 int integers
float64 float floats
bool bool booleans
datetime64[ns] datetime.datetime date/time
timedelta64[ns] datetime.timedelta periods of time
object mixed, typically str Mixed types (typically str)
Topics not covered:
Advantages of pandas relative to Excel/lists/dicts (Lec 5.1)
Lecture 6: Accessing data in Pandas: Indexing and I/O (I)
Important topics include:
How to use .loc and .iloc
And what [] does
For .loc and iloc, which objects are returned by obj.loc[indexer],
obj.iloc[indexer], and obj[indexer]
When obj is either a series or a data frame
When indexer is either a scalar, list, or slice (series, data frames)
When indexer is a combination of row and column indexers (data frames)
e.g., df.loc[:, indexer], df.iloc[row_indexer, col_indexer], . . .
Lecture 6: Accessing data in Pandas: Indexing and I/O (II)
Important topics include (cont):
Make sure you know what these methods do:
obj.set_index, obj.reset_index
pandas.read_csv (and the parameters index_col and parse_dates)
obj.to_csv (no need to know what optional parms do)
Lecture 7: Working with time-series (I)
Important topics include:
The datetime module
datetime.datetime
Attributes day, month, year, hour, minute, second, microsecond
Methods .now and .strftime (no need to memorise the directives)
datetime.timedelta
Remember that seconds is not total seconds if > 1 day
DatetimeIndex objects
How to construct them (e.g., pandas.to_datetime)
How to compute returns (the .pct_change method)
No need to use the freq parameter
How to select observations (df.loc[‘2020’] → all obs for the year 2020)
Lecture 7: Working with time-series (II)
Topics not included:
Advantages of DatetimeIndex objects (in Lec 7.2)
Lecture 8: Event studies in Finance
Important topics include:
What is an event study?
CAR as an outcome variable, including
What is CAR trying to measure?
Why is CAR computed over a window around the event date?
How to test a null hypothesis (i.e, how to compute a t-stat)
The complete example (in Lec 8.2)
Lecture 9: Doing more with Pandas (I)
Important topics include:
Joining pandas objects
The .join method (and the parameter how)
Selection using booleans:
How to create boolean arrays (e.g, df.loc[:, col] == value)
How to use obj.loc[cond], where cond is a boolean array
How to combine conditions using | and &
No need to know the difference between | and or (or & and and)
Make sure you know what df.isna() does
Lecture 9: Doing more with Pandas (II)
Important topics include (continued):
Split-apply-combine operations using Pandas (“groupby”)
How to create GroupBy objects
The attribute GroupBy.groups
Make sure you understand the following GroupBy methods:
.last, .count, .size
How the method obj.apply works, when obj is either
A GroupBy object, a series, or a data frame
For example, what we did in week9_slides_XXX.py
The lec_utils.py module
Developed to help us visualise objects and create fake CSV files
You only need to know what the following methods do:
lec_utils.pprint(obj) → Pretty-prints object
lec_utils.csv_to_fobj(cnts) → Creates strings that behave like files
lec_utils.csv_to_df(cnts) → Creates a DF (just like pd.read_csv
but accepts CSV-formatted strings)
No need to understand how it works
There will be no question asking you to explain how it works
Will only be used to create test data frames and to print stuff
See the practice final exam for an example
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com