₩228,000-4.8%vs prev closeclose of 2026-10-30 · not real-time
Valuation summary
Looks expensive
GAON CABLE Co., Ltd., a cable company, provides industrial power cables in South Korea. The company offers power cables, including extra-high voltage cables, such as XLPE insulation cables; overhead lines comprising overhead transmission lines, optical fiber composite overhead…· Source: company profile (yfinance/Yahoo Finance)
vs. analyst consensus target. The target is an opinion, not a prediction or buy signal.
Score summary
A summary of valuation factors — not a buy recommendation. You decide.
🕒Data as of: 2026-10-30 · end-of-day, not real-timeData as of — price/valuation/technicals: the close date above · NPS stake: US=quarterly SEC 13F, Korea=annual · short interest: exchange's latest settlement · fundamentals: latest filing
The chart (long-term trend, Ichimoku cloud) is for observation only — not a buy or sell signal.
Structure analysis · observation only
Shows only where the current price sits within the channel structure (based on past swings).
Position: 0.0~0.5 zone · Uptrend channelStructure as of 2026-10-30
Channel levels (price by multiple)
×3₩659,148
×2.5₩586,395
×2₩513,643
×1.5₩440,891
×1₩368,138
×0.5₩295,386
◆Current price₩228,000
Baseline (0)₩222,634
Unfilled gaps
Gap downabove price₩231,500–₩232,500 · Gap width 0.43% · 2026-10-30
Gap downabove price₩273,500–₩275,000 · Gap width 0.55% · 2026-10-15
Gap upbelow price₩105,667–₩109,333 · Gap width 3.5% · 2026-04-23
Gap upbelow price₩59,667–₩59,889 · Gap width 0.37% · 2026-04-15
Gap upbelow price₩46,406–₩47,128 · Gap width 1.6% · 2026-02-09
Gap upbelow priceBeyond chart₩27,616–₩28,810 · Gap width 4.3% · 2025-04-23
Gap upbelow priceBeyond chart₩22,093–₩22,176 · Gap width 0.38% · 2025-04-14
Gap upbelow priceBeyond chart₩20,122–₩20,677 · Gap width 2.8% · 2025-04-10
The channel, levels and gaps are observational geometry showing where the current price sits within past swing structure. They are not a rebound/bottom probability, a price target, or a buy/sell signal — plain chart patterns carry no reliable predictive edge (our backtest conclusion).
A highlight of the key metrics detailed below.
Undervalued vs sector 21/100 within Technology — higher means lower multiples (PER·PB·EV/EBITDA·PSR) than sector peers. Growth sectors carry higher multiples, so this is a same-sector relative view, not absolute cheapness (observational).
Why it looks cheap
–
Why it might be risky
P/E (Price/Earnings)104.0
Forward P/E33.4
P/B (Price/Book)12.6
Strategy ranks
Top-tier in 0 of 7 evaluated strategies 0/7
Ranks within the Korea market — tap to open that ranking (ranks are per market).
Past returns — they do not guarantee future results.
More indicators
Sortino
3.74
Downside vol
75.65%
Max drawdown (3y)
-65.66%
Deep historical drawdown
Volatility (annual.)
133.12%
Large price swings
RSI(14) daily
41.80
RSI(14) weekly
52.70
RSI(14) monthly
63.20
vs 200d MA
46.25%
Above 200-day avg
Trend slope (annualized)
159.19%
Recent uptrend
FCF margin
-7.90%
Beta
2.57
Moves more than market
Payout ratio
2.14%
Sustainable
Price/Graham
7.63
Pricey vs assets/earnings
Golden cross
○ aligned
Color: green = healthy, red = caution (absolute thresholds, value/quality metrics only). Downside vol & Sortino are annualized. Max drawdown = largest drop from a 3-year peak; volatility = 1-year annualized. Price/Graham < 1 = below Graham fair value (profitable, positive-equity names only). RSI < 30 = oversold. For reference only.
Long-term trend (Ichimoku)
A Japanese trend tool. Price above the 'cloud' is supportive; below is weak. A strong uptrend doesn't mean it's cheap — stocks that have run up are usually more expensive.
D
Inside the cloud (neutral)
W
Above the cloud (supportive)
M
Above the cloud (supportive)
RSI divergence (timeframes)
Daily · cleared gate (tiny effect)
Of 24 backtested chart patterns, only the daily bullish variant cleared the 20-day excess-return gate (t=2.23). The excess was +0.68% — economically negligible — and the bearish variant did not clear it. Treat it as an observation, not a probability or buy signal.
–
Weekly · observational
Observational — no predictive edge confirmed in backtest (not a signal).
–
Monthly · observational
Observational — no predictive edge confirmed in backtest (not a signal). Over long holds it has tended to underperform the average.
📉 Bearish divergence (observed)
High pivots
2026-04-30 → 2026-09-30
When price falls while RSI makes a higher low, you are 'observing' selling pressure fading — not a forecast or a buy signal.