Unit 1: Financial data and time series
Financial Analytics notes · PTU syllabus (MBA 965-26)
On this page
Unit summary
Finance runs on data, and analytics turns prices, transactions and statements into decisions. This unit covers the role of analytics in finance, types of financial data, data wrangling and cleaning, exploratory data analysis, time series components and decomposition, stationarity, ACF and PACF, and an introduction to ARMA and ARIMA models.
After this unit you can
- Explain the role of analytics in finance and types of financial data
- Wrangle, clean and explore financial data
- Decompose financial time series and test stationarity
- Use ACF, PACF and ARMA/ARIMA models
PTU syllabus topics
- Role of analytics in finance
- types of financial data
- data wrangling and cleaning
- exploratory data analysis
- time series components
- decomposition
- stationarity
- ACF/PACF
- introduction to ARMA/ARIMA
- 1
Clean price data
- 2
Compute returns
Log or simple
- 3
Explore
Plots, summary stats
- 4
Test stationarity
ADF test
- 5
Fit ARMA or ARIMA
- 6
Check residuals
Topic 1
Role of analytics in finance
- Applications: investment research and portfolio management, risk measurement (market, credit, liquidity), fraud detection, credit scoring, algorithmic trading, financial planning and forecasting, regulatory reporting, customer analytics in banking.
- Drivers: abundant market and transaction data, computing power, regulation (Basel, IFRS 9 expected credit loss), competition from fintechs.
Topic 2
Types of financial data
Market data
Prices, returns, volumes, order books, yields, exchange rates
Fundamental data
Financial statements, ratios, earnings
Transaction data
Payments, card swipes, UPI transactions
Economic data
GDP, inflation, interest rates
Alternative data
News, social media sentiment, satellite images, web traffic
- Frequency: tick, intraday, daily, monthly, quarterly; sources: NSE and BSE, RBI DBIE, CMIE Prowess, Bloomberg, Refinitiv, company filings.
Topic 3
Data wrangling and cleaning
- Common issues: missing prices (holidays), corporate actions (splits, bonuses, dividends) requiring adjusted prices, outliers and errors, different time zones and calendars, survivorship bias (delisted stocks missing).
- Steps: import, align dates, adjust for corporate actions, handle missing values (forward fill, interpolation), compute returns, check outliers.
Simple return
(Pt − Pt−1) ÷ Pt−1
Log return
ln(Pt ÷ Pt−1)
Annualised volatility
Daily standard deviation × √252
Topic 4
Exploratory data analysis
- Summary statistics: mean, standard deviation, skewness, kurtosis of returns — financial returns are often fat-tailed (excess kurtosis) and negatively skewed.
- Visual tools: price and return charts, histograms with normal overlay, Q–Q plots, rolling volatility, correlation heat maps.
Topic 5
Time series components and decomposition
- Components: trend, seasonality, cycles and irregular movements; financial prices usually show trends and random fluctuations, returns are closer to random.
- Decomposition: additive or multiplicative; moving-average and STL decomposition.
Topic 6
Stationarity, ACF and PACF
- Prices are usually non-stationary (random walk); returns are approximately stationary.
- Tests: Augmented Dickey–Fuller and KPSS tests.
- ACF and PACF identify dependence in returns or volatility; significant ACF in squared returns indicates volatility clustering.
Topic 7
Introduction to ARMA and ARIMA
AR(p)
Value depends on its own past values: Yt = c + φ1 Yt−1 + … + et
MA(q)
Value depends on past errors: Yt = μ + et + θ1 et−1 + …
ARMA(p, q)
Both components for stationary series
ARIMA(p, d, q)
ARMA after d differences
SARIMA
Adds seasonal terms
AR(p)
Tails off gradually
Cuts off after lag p
MA(q)
Cuts off after lag q
Tails off gradually
ARMA
Tails off
Tails off
- In finance: returns often show weak autocorrelation, so ARMA forecasts of returns are modest; ARIMA is more useful for macro and accounting series (sales, interest rates).
Key terms
- Alternative data
- Non-traditional data such as sentiment or satellite images
- Adjusted price
- Price corrected for splits and dividends
- Log return
- Natural log of the price ratio
- Fat tails
- More extreme outcomes than the normal distribution
- Stationarity
- Constant mean and variance over time
Quick revision
- Uses of analytics in finance and drivers.
- Market, fundamental, transaction, economic, alternative data; sources.
- Wrangling issues and steps; simple and log returns; annualised volatility.
- EDA; fat tails; decomposition.
- Prices non-stationary, returns stationary; ADF; ACF/PACF; ARMA, ARIMA.
Important exam questions
Practice questions written to the PTU exam pattern for this unit's syllabus: short answers (Section A style) and long answers (Sections B and C style).
Short-answer questions
- Q1.State three applications of analytics in finance.
- Q2.What is alternative data?
- Q3.Why are prices adjusted for corporate actions?
- Q4.Distinguish simple and log returns.
- Q5.Why are returns used instead of prices in modelling?
- Q6.What do ACF spikes in squared returns indicate?
Long-answer questions
- Q1.Explain the role of analytics in finance and types of financial data.
- Q2.Explain data wrangling and exploratory analysis of financial data.
- Q3.Explain time series decomposition and stationarity in financial data.
- Q4.Explain ACF, PACF and ARMA/ARIMA models for financial series.
Stuck on this unit?
Message SBS on WhatsApp for help with Financial Analytics, or to ask about studying MBA at Synetic.