Frequently Asked Questions

NEPSE's continuous trading session runs from 11:00 AM to 3:00 PM Nepal Time (NPT, UTC+5:45). That's the window where regular buy / sell orders are matched on the floor.

• 10:30 – 11:00 AM — Pre-open session (order entry only, no matching)
• 11:00 AM – 3:00 PM — Continuous trading
• 3:00 – 3:10 PM — Post-close session

Sunday through Thursday. NEPSE follows the Nepali working week — Friday is a half-holiday for most institutions and Saturday is the weekly public holiday, so the exchange stays closed both days.

Public holidays declared by the government also close the market — see the "holidays" question below.

Inside the trading window, the backend polls upstream sources roughly every 60 seconds and upserts prices, indices, top gainers/losers, and floorsheet trades. Pages built on React Query refetch on the same cadence so what you see is at most ~1 minute behind the floor.

Outside trading hours the sync loop sleeps and the dashboard shows the last available close.

The "open / closed" badge isn't just a clock check. Two sources agree:

• Local schedule — Sun–Thu, 10:30 – 15:15 NPT
• Upstream NEPSE flag — theisOpen field returned by NEPSE's own status API

On a public holiday, an unscheduled half-day, or a tech outage, NEPSE will report closed even though the clock says trading hours — and we honor that.

No. NEPSE follows the Nepal Government calendar — Dashain, Tihar, Chhath, May Day, Constitution Day, and many regional festivals all close the market. Some of these are multi-day closures (Dashain is typically a full week).

The exchange publishes its annual holiday list on nepalstock.com; the platform picks up the daily isOpen flag automatically so the dashboard reflects holidays without manual updates.

A daily price band that limits how far an individual stock can move from its previous close. NEPSE applies a ±10% circuit on most ordinary shares — once a stock hits that ceiling or floor, trading in it pauses for the session.

Bonus shares, rights issues, and IPO listings have different bands on their first trading day.

NEPSE Index covers all listed ordinary shares — hydropower, micro-finance, banks, hotels, every sector.

Sensitive Index is a curated subset: only "Class A" companies (commercial banks, large insurance, blue-chip names). It moves slower and is treated as a quality benchmark.

We use a K-Means clustering model that looks at 5 features for each broker on a given stock:

• Average trade size — how big a typical order is
• Market share — % of the day's volume they handle
• Trade count — how often they show up
• Direction consistency — do they keep buying or flip-flop?
• Volume rank — where they rank vs others on this stock

Market share alone doesn't decide tier — a broker with low share but high trade count and consistent direction can still be Tier 1.

Three steps:

Score = Σ (buy_ratio × volume) ÷ Σ (volume)

Δ = Score(today) − Score(10 days ago)

A BUY or SELL can be downgraded to HOLD if the broker math says one thing but the price chart disagrees. We check three filters:

• Volume confirmed — today's volume isn't unusually low compared to the 21-day average
• Not overextended — the stock hasn't already run up too far recently
• Tier-1 ≥ 2 — at least 2 big brokers agree (one broker isn't enough)

The right-side "Today's Broker Signal Snapshot" is computed once per day by a scheduled job at 16:00 NPT (post-close). The hero card on the left re-computes live on every page load.

Between the snapshot run and now, more floorsheet trades for today may have been upserted. A small score drift (e.g., 0.596 → 0.53) can flip BUY → HOLD because there's no hysteresis on the threshold (BUY needs score > 0.55).

When this happens, the hero shows an amber pill explaining the divergence. Both numbers are correct — they're just snapshots of the data at two different times.

Confidence reflects how many Tier-1 brokers were active on this stock today:

• High — 3 or more active
• Medium — 2 active
• Low — fewer than 2 (signal is downgraded to HOLD)

Low confidence = not enough institutional voices to form a meaningful read. The signal forces HOLD as a safety floor when this happens.

Once a BUY signal fires, we suggest where to enter, where to cut losses, and where to take profit — sized to the stock's actual recent volatility (14-day ATR), not a flat percentage.

Entry       = today's close
Stop Loss   = Entry − (multiplier × ATR₁₄)
Take Profit = Entry + (multiplier × ATR₁₄)

Stronger signals (STRONG_BUY) get tighter stops and bigger targets.

By design — they only make sense for actionable, long-direction trades:

• HOLD — no trade is being suggested, so no entry / stop / target.
• SELL / STRONG_SELL — NEPSE doesn't allow retail short-selling, so a SELL means "exit your existing position", not "open a short". The exit price IS the signal — no additional levels needed.

R:R = Risk : Reward — how much you stand to gain vs. how much you'd lose if the trade goes wrong.

The system requires at least 1.5 : 1 before recommending the trade. The reasoning is statistical:

• At 1.5:1 R:R, you can be wrong 40% of the time and still break even.
• At 1.25:1 R:R, you need to be right ~45% of the time just to break even.
• After commissions and slippage, anything under 1.5:1 is mathematically a coin flip.