MA2830 — Portfolio Hedging (Tableau Assignment)

Finance Crash Course (Start Here)

Portfolio instrument: <PORTFOLIO_TICKER>
Hedge instrument:     <HEDGE_TICKER>

Upfront Assessment (Exploratory Visuals)

Upfront EDA — Guiding Questions (answer in captions)

Get Live Data (choose one)

Portfolio instrument: <PORTFOLIO_TICKER>
Hedge instrument:     <HEDGE_TICKER>

A. Google Sheets (no code)

1. Create tabs: PORT, HEDGE, MERGE. Put tickers in PORT!A1 and HEDGE!A1.
2. In PORT!A3 paste:

   Copy=QUERY(GOOGLEFINANCE($A$1,"close",EDATE(TODAY(),-36),TODAY()),
    "select Col1,Col2 where Col2 is not null label Col1 'Date', Col2 'Close'",1)

3. Repeat same in HEDGE!A3.
4. In MERGE!A2 paste:

   Copy=QUERY({PORT!A3:B, ARRAYFORMULA(IFNA(VLOOKUP(PORT!A3:A, HEDGE!A3:B, 2, FALSE)))},
    "select Col1,Col2,Col3 where Col3 is not null label Col1 'Date', Col2 'PortfolioClose', Col3 'HedgeClose'",0)

   Download MERGE as prices_clean.csv.

B. Python (yfinance, ~10 lines)

Copyimport yfinance as yf, pandas as pd
PORT, HEDGE = "QQQ", "SPY"
end = pd.Timestamp.today().normalize(); start = end - pd.DateOffset(months=36)
def pull(t):
    df = yf.download(t, start=start, end=end, interval="1d", auto_adjust=True)[["Close"]]
    return df.reset_index().rename(columns={"Close":t})
df = pull(PORT).merge(pull(HEDGE), on="Date", how="inner")
df = df.rename(columns={PORT:"PortfolioClose", HEDGE:"HedgeClose"})
df.to_csv("prices_clean.csv", index=False)

Tableau Steps (calculated fields)

1. Portfolio Return: ( SUM([PortfolioClose]) / LOOKUP(SUM([PortfolioClose]), -1) ) - 1
2. Hedge Return: ( SUM([HedgeClose]) / LOOKUP(SUM([HedgeClose]), -1) ) - 1
3. Parameter: Hedge Ratio (0.00–1.50, step 0.10)
4. Hedged Return (Static): [Portfolio Return] - [Hedge Ratio] * [Hedge Return]
5. Rolling Beta (90d): WINDOW_COVAR([Portfolio Return],[Hedge Return], -89, 0) / WINDOW_VAR([Hedge Return], -89, 0)
6. Beta Clamped: MIN(1.5, MAX(0, [Rolling Beta (90d)]))
7. Hedged Return (Dynamic): [Portfolio Return] - [Beta Clamped] * [Hedge Return]
8. Equity (per series): 100 * EXP(RUNNING_SUM(LOG(1+[Return Series])))
9. Drawdown (per series): [Equity] / WINDOW_MAX([Equity]) - 1
10. KPIs (per series): Ann Return, Ann Vol, Max DD, Sharpe = Return/Vol. Compute Using: Date.

Implementation (Tableau-ready)

Copy[Portfolio Return]  = ( SUM([PortfolioClose]) / LOOKUP(SUM([PortfolioClose]),  -1) ) - 1
[Hedge Return]      = ( SUM([HedgeClose])     / LOOKUP(SUM([HedgeClose]),     -1) ) - 1

Copy[Rolling Beta (90d)] =
WINDOW_COVAR( [Portfolio Return],  [Hedge Return], -89, 0 )
/
WINDOW_VAR ( [Hedge Return],                  -89, 0 )

Copy[Hedged Return (Static)]  = [Portfolio Return] - [Hedge Ratio] * [Hedge Return]
[Hedged Return (Dynamic)] = [Portfolio Return] - [Rolling Beta (90d)] * [Hedge Return]

CopyEquity = 100 * EXP( RUNNING_SUM( LOG( 1 + [Return Series] ) ) )

Deliverables

• Upfront EDA Dashboard — Monthly Return Heatmap, Box Plot by Month, Histogram, Rolling 90-day Correlation Heatmap (captions answer EDA questions).
• Tableau Storyboard (5–7 story points = 5–7 dashboards). Set Story as default tab; turn off “Show Sheets as Tabs”. Export a PowerPoint (.pptx) of the Story.
• Required Dashboards (used as story points):
  • D1 — Overview — Equity curves: Unhedged vs Static vs Dynamic; Hedge Ratio visible; portfolio & hedge tickers shown in title/subtitle.
  • D2 — Rolling Beta — 90-day beta line + callouts when beta < 0.3 or > 1.2
  • D3 — Static vs Dynamic — KPIs table + one example where dynamic helps/hurts
  • D4 — Sensitivity (Static) — ratios 0.5, 0.8, 1.0, 1.2 (small multiples or parameter view)
  • D5 — Stress Windows — pick 2–3 shock windows; per-window bars or mini-curves
  • D6 — Decision — one key chart + 3 bullets: Benefit, Cost, Limitation
  • D7 — Optional — Worst 5-day loss per strategy (bar chart)
• Workbook — .twbx (Tableau Packaged Workbook) ONLY; no Tableau Public links. The .twbx must contain all dashboards and the Story and open offline.
• Instrument disclosure — In your written report (Data & Instruments section) and in D1 Overview, clearly state Portfolio instrument and Hedge instrument using the exact ticker symbols you used in your data pull.
• Data file — prices_clean.csv that:
  • contains at least Date, PortfolioClose, HedgeClose, and
  • includes explicit ticker identifiers in either:
    • the column names themselves (e.g. META_Close, QQQ_Close), or
    • additional string columns such as PortfolioTicker and HedgeTicker repeated on each row.
  The tickers in this CSV MUST match the tickers stated in your write-up and shown in your Tableau story. Submissions without clear ticker identifiers in the CSV will receive a deduction.
• Write-up — 1,500–2,000 words covering context, definitions, method, KPIs, sensitivity, stress windows, decision, and ≥3 credible references.
  
  Your write-up MUST include a “Data & Instruments” paragraph in the following format (with all placeholders filled in):

  Data and Instruments. In this project I hedge a long position in <PORTFOLIO_NAME> (ticker: <PORTFOLIO_TICKER>)
  using <HEDGE_NAME> (ticker: <HEDGE_TICKER>) as the hedge instrument.
  Daily adjusted closing prices for both tickers were obtained from <DATA_SOURCE> over the period
  <START_DATE> to <END_DATE>, downloaded using <TOOL_USED: e.g. Google Sheets GOOGLEFINANCE /
  Python yfinance> and cleaned into a single file prices_clean.csv with columns Date,
  PortfolioClose and HedgeClose (plus optional helper fields such as Date (Months) and Stress Window).
  The portfolio series represents the asset I am trying to protect, while the hedge series represents the
  instrument used to reduce risk; these exact tickers are used consistently in my data pipeline,
  Tableau workbook and dashboards (D1 Overview in particular).

Grading Rubric (100 pts)