OpenEDG PCAD — Certified Associate Data Analyst with Python (PCAD-31-02)

Given `df = pd.read_csv('sales.csv')`, which method returns the first 5 rows by default?

• Adf.first()
• Bdf.head()
• Cdf.top(5)
• Ddf.peek()

Which Pandas function loads a CSV file from disk into a DataFrame?

• Apd.load_csv
• Bpd.read_csv
• Cpd.open_csv
• Dpd.csv

What does `df.shape` return?

• ANumber of rows only
• BA tuple `(n_rows, n_columns)`
• CMemory size in MB
• DColumn names

Which method gives summary statistics (mean, std, min, max, quartiles) for numeric columns?

• Adf.summary()
• Bdf.describe()
• Cdf.stats()
• Ddf.info()

How do you select rows where the `age` column is greater than 30?

• Adf[df['age'] > 30]
• Bdf.where(age > 30)
• Cdf.select(age > 30)
• Ddf.filter('age > 30')

What is the difference between `df.loc[]` and `df.iloc[]`?

• AThey are identical
• B`loc` uses labels (row/column names); `iloc` uses integer positions
• C`loc` is deprecated
• D`iloc` works only on Series

Given `s = pd.Series([1, 2, 3, 4, 5], index=['a','b','c','d','e'])`, what does `s.iloc[1:3]` return?

• AThe values at positions 1 and 2 (i.e., 2 and 3, with labels 'b' and 'c')
• BThe values at labels 1 and 3
• CAn empty series
• DAll five values

Which method drops rows containing any NaN values?

• Adf.dropna()
• Bdf.fillna()
• Cdf.isna()
• Ddf.removena()

How do you fill all NaN values in a numeric DataFrame with the column mean?

• Adf.fillna(df.mean())
• Bdf.replace(NaN, mean)
• Cdf.dropna(fill='mean')
• Ddf.mean(fillna=True)

Which method removes duplicate rows from a DataFrame?

• Adf.unique()
• Bdf.distinct()
• Cdf.drop_duplicates()
• Ddf.dedup()

Given two DataFrames `customers` and `orders` both with a `customer_id` column, how do you perform an inner join on that column?

• Apd.join(customers, orders)
• Bpd.merge(customers, orders, on='customer_id', how='inner')
• Ccustomers + orders
• Dcustomers.combine(orders)

What does `df.groupby('region')['sales'].sum()` return?

• AA DataFrame with all columns summed
• BA Series with the sum of `sales` per `region`
• CA list of regions
• DAn error

Which is the correct syntax for applying multiple aggregations to a single column in groupby?

• Adf.groupby('cat')['val'].agg(['sum', 'mean', 'std'])
• Bdf.groupby('cat')['val'].apply(['sum'])
• Cdf.groupby('cat').sum_mean()
• Ddf.groupby('cat')['val'].multi('sum','mean')

What does `transform` do in a groupby?

• AReturns one value per group
• BReturns a result aligned to the original DataFrame's index — same shape — useful for comp…
• CDrops all groups
• DSorts the DataFrame

How do you reshape a DataFrame from long to wide format?

• Adf.pivot(index='date', columns='product', values='sales')
• Bdf.melt()
• Cdf.transpose()
• Ddf.reshape(wide=True)

What does `pd.melt(df, id_vars=['date'], value_vars=['a','b','c'])` do?

• ADrops the columns a, b, c
• BReshapes wide → long: keeps `date` as identifier and stacks columns a, b, c into two new…
• CComputes the mean of a, b, c
• DAggregates a, b, c

Which NumPy function creates an array of evenly-spaced values from 0 to 9 inclusive?

• Anp.range(0, 10)
• Bnp.arange(10)
• Cnp.linspace(0, 9, 1)
• Dnp.zeros(10)

What does `arr.shape` return for `arr = np.array([[1,2,3],[4,5,6]])`?

• A(6,)
• B(2, 3)
• C(3, 2)
• D6

What does `arr.reshape(3, -1)` do for a 12-element array?

• AErrors out
• BReshapes to (3, 4) — the -1 means 'infer this dimension'
• CReshapes to (3, 12)
• DReverses the array

What is broadcasting in NumPy?

• AA network protocol
• BA set of rules for performing element-wise operations on arrays of different but compatib…
• CA way to share arrays between processes
• DA streaming API

Which call computes the column-wise mean of a 2D NumPy array `arr`?

• Aarr.mean()
• Barr.mean(axis=0)
• Carr.mean(axis=1)
• Dnp.col_mean(arr)

What does `np.where(arr > 0, 1, -1)` return?

• AJust the values greater than 0
• BAn array the same shape as `arr` with 1 where `arr > 0` and -1 elsewhere
• CA boolean mask
• DAn error

Which statement creates a Pandas datetime index from `df['date']` and sets it as the index?

• Adf.index = df['date']
• Bdf['date'] = pd.to_datetime(df['date']); df = df.set_index('date')
• Cdf.dateindex = df['date']
• Ddf.set_date_index()

Given a DatetimeIndex, what does `df.resample('M').sum()` do?

• ASums all rows
• BResamples to monthly frequency, summing values within each month
• CFilters to rows in May only
• DErrors out

Which `.dt` accessor extracts the day of week (0=Monday) from a datetime Series?

• As.dt.weekday
• Bs.dt.day
• Cs.dt.weekofyear
• Ds.dt.dow

Which scikit-learn function splits arrays/matrices into train and test subsets?

• Asklearn.split.train_test()
• Bfrom sklearn.model_selection import train_test_split
• Csklearn.preprocess.split()
• Dsklearn.utils.partition()

Which method does a scikit-learn estimator use to learn parameters from training data?

• Aestimator.predict(X)
• Bestimator.train(X, y)
• Cestimator.fit(X, y)
• Destimator.learn(X, y)

Which is the correct order of operations to standardize features before training a model?

• Ascaler.fit_transform on the entire dataset, then train_test_split
• Btrain_test_split first, then scaler.fit_transform on X_train and scaler.transform on X_te…
• CSkip scaling for tree-based models — neither order matters
• Dscaler.transform on X_train and scaler.fit on X_test

In a binary classification problem with class imbalance (95% negative, 5% positive), which metric is most misleading?

• APrecision
• BRecall
• CAccuracy
• DROC AUC

What does `sklearn.metrics.confusion_matrix(y_true, y_pred)` return for binary classification?

• AA scalar accuracy score
• BA 2x2 array `[[TN, FP],[FN, TP]]`
• CA list of class labels
• DA 4x4 matrix

For a regression model, which metric is the square root of the average squared residual?

• AMAE
• BMSE
• CRMSE
• DR²

What does `sklearn.linear_model.LinearRegression().fit(X, y).coef_` return?

• APredicted values
• BThe fitted regression coefficients (slopes) per feature
• CThe intercept
• DResiduals

What is the primary purpose of `KMeans` in scikit-learn?

• ASupervised regression
• BUnsupervised clustering — partition observations into k clusters by minimizing within-clu…
• CDimensionality reduction
• DClassification with labels

Which technique reduces dimensionality by projecting data onto orthogonal axes of maximum variance?

• APCA (Principal Component Analysis)
• BKMeans
• CLogistic Regression
• DTrain-test split

What does `sklearn.model_selection.cross_val_score(model, X, y, cv=5)` do?

• ATrains the model 5 times on the entire dataset
• BPerforms 5-fold cross-validation: trains on 4 folds, evaluates on the held-out fold, repe…
• CSplits data 5/95
• DOptimizes hyperparameters

Which scikit-learn class searches over a hyperparameter grid using cross-validation?

• AGridSearchCV
• BRandomCV
• CHyperTune
• DParamSearch

Which Pandas method writes a DataFrame to a SQL database?

• Adf.write_sql()
• Bdf.to_sql()
• Cdf.export_sql()
• Ddf.save_sql()

Which is a parameterized SQL query (safe from SQL injection) using `sqlite3`?

• Acursor.execute(f"SELECT * FROM users WHERE id = {user_id}")
• Bcursor.execute("SELECT * FROM users WHERE id = ?", (user_id,))
• Ccursor.execute("SELECT * FROM users WHERE id = " + str(user_id))
• Dcursor.execute("SELECT * FROM users WHERE id = %s" % user_id)

Which Matplotlib call creates a histogram of data in array `x`?

• Aplt.bar(x)
• Bplt.hist(x, bins=30)
• Cplt.scatter(x, x)
• Dplt.line(x)

What does `seaborn.heatmap(corr, annot=True, cmap='coolwarm')` typically display?

• AA scatter matrix
• BA correlation matrix as a colored grid with numerical values overlaid
• CA box plot
• DA 3D surface

Which is the typical purpose of a box plot in EDA?

• AShow categorical frequencies
• BVisualize the distribution of a numeric variable: median, IQR (Q1 to Q3), whiskers, and o…
• CPlot a regression line
• DShow 3D surfaces

Which Python statement opens a JSON file for reading?

• Ajson.read('file.json')
• Bwith open('file.json') as f: data = json.load(f)
• Cjson.parse('file.json')
• Dopen.json('file.json')

What is the difference between `json.load()` and `json.loads()`?

• ANo difference
• B`load(f)` reads from a file object; `loads(s)` parses a JSON string
• C`load` is for arrays; `loads` is for objects
• D`loads` is deprecated

Which is a try/except pattern for safely parsing user-supplied integers from a CSV column?

• Atry: x = int(row['v']) except ValueError: x = None
• Bx = int(row['v']) — no exception handling needed
• Ctry: x = int(row['v']) except: pass
• Dif isinstance(row['v'], int): x = row['v']

Which Pandas option returns a copy with a column renamed?

• Adf.rename(columns={'old':'new'})
• Bdf.column_rename('old', 'new')
• Cdf['old'] = 'new'
• Ddf.alter('old', 'new')

What does `df.dtypes` return?

• AA DataFrame summary
• BA Series mapping each column name to its NumPy dtype (e.g., int64, float64, object)
• CMemory usage
• DColumn count