Trading vs Investing & Styles

Q1 (10 marks) — Trader vs Investor mindset

Award 1 mark each side (2 per dimension):

Full 2 only if both columns correct for a dimension.

Q2 (12 marks) — Transaction cost derivation

(a) (4 marks) Cost per round trip $= c \cdot K$. Trades per month $= n \cdot d$. $\text{MonthlyCost} = n \cdot d \cdot c \cdot K$ (1 for cost/trip, 1 for trades/month, 2 for combined formula.)

(b) (3 marks) $n=8,\; d=20,\; c=0.0005,\; K=500000$. $= 8 \times 20 \times 0.0005 \times 500000 = 160 \times 0.0005 \times 500000$ $= 160 \times 250 = ₹40{,}000$ (1 sub, 1 arithmetic, 1 answer ₹40,000.)

(c) (5 marks) Cost as % of capital $= 40000/500000 = 8\%$. Net return $=$ Gross $-$ cost%. $\text{Gross} = \text{Net} + \text{Cost\%} = 3\% + 8\% = 11\%$ (2 for cost as % of capital, 2 for relation, 1 for answer $11\%$.) Teaching note: high trade frequency makes costs the dominant hurdle — this is why scalpers need very cheap execution.

Q3 (10 marks) — Style decision table

Table (7 marks — ~0.5/cell, round up):

Justification (3 marks): Scalping extracts many tiny edges from a small capital base within one session, so capital is recycled many times per hour → high capital-efficiency-per-hour (1). But the continuous vigilance, split-second decisions, and rapid P&L swings create maximal cognitive load and stress (1); a single lapse erases many wins, so the risk-per-mistake is severe (1).

Q4 (10 marks) — Mean-reversion code

Code (8 marks):

def mean_reversion_signals(prices, w=20, k=2.0):
    signals = []
    for i in range(len(prices)):
        if i < w - 1:
            signals.append("HOLD")          # not enough data
            continue
        window = prices[i-w+1 : i+1]
        mean = sum(window) / w
        var = sum((x - mean)**2 for x in window) / w
        std = var ** 0.5
        z = (prices[i] - mean) / std if std > 0 else 0
        if z < -k:
            signals.append("BUY")           # price too low -> revert up
        elif z > k:
            signals.append("SELL")          # price too high -> revert down
        else:
            signals.append("HOLD")
    return signals

Marks: rolling window (2), mean+std (2), z-score (2), threshold logic BUY/SELL/HOLD (2).

$k$ = number of standard deviations from the mean (the z-threshold) that defines "extreme enough to trade."

Contrast (2 marks): Momentum does the opposite — it BUYs strength ($z>+k$ / rising) and SELLs weakness, betting the move continues, whereas mean-reversion bets the extreme reverses.

Q5 (10 marks) — Return expectations

(a) Three reasons (6 marks, 2 each):

1. Compounding math is absurd — 10%/day for 20 days compounds to unrealistic multiples; no consistent edge survives at that magnitude.
2. Transaction costs & slippage eat scalping profits (see Q2) — high frequency means costs dominate; realistic net edges are tiny per trade.
3. Realistic return expectations — even skilled professionals target a few % per month; win-rate and risk-of-ruin make daily double-digit returns statistically impossible to sustain (drawdowns wipe accounts).

(b) (4 marks): Compounded annual $= (1+0.02)^{12} - 1$. $(1.02)^{12} = 1.26824\ldots \Rightarrow \approx 26.8\%$ (2 for compounding formula, 1 computation, 1 answer $\approx 26.8\%$.) Note: still strong and realistic vs the 200% claim.

Q6 (8 marks) — Personality fit

Recommendation (2 marks): Swing trading (or lean positional). Fits limited screen-time (checks twice/day), slower deliberate decisions, multi-day holds, and modest capital works.

Justify rejections (6 marks, 2 each):

• Scalping — rejected: requires continuous screen presence and instant decisions; contradicts full-time job + dislike of fast decisions.
• Intraday — rejected: needs multi-hour active monitoring and must flat by close; impossible with only twice-daily checks.
• Positional (if not chosen) or momentum-scalp — acceptable alternative to positional; positional also fits but swing better matches a "weeks" horizon with ₹2L capital and periodic checking; reject aggressive momentum-intraday for the same time-constraint reason.

(Accept positional as the recommendation with equivalent justification; grade on internal consistency.)

[
  {"claim":"Monthly cost = 8*20*0.0005*500000 = 40000","code":"result = (8*20*0.0005*500000 == 40000)"},
  {"claim":"Cost as percent of capital = 8%","code":"result = (40000/500000 == 0.08)"},
  {"claim":"Required gross monthly return = 3% + 8% = 11%","code":"result = (0.03 + 0.08 == 0.11)"},
  {"claim":"Compounded annual from 2% monthly is approx 26.82%","code":"val=(1+Rational(2,100))**12 - 1; result = abs(float(val)-0.2682)<0.001"}
]