company-valuation

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Company Valuation

公司估值

Triangulates intrinsic value via three methods, then blends them to an implied share price:

DCF — 5-year FCFF projection, discount at WACC, terminal value.
Relative — apply peer median P/E, EV/Revenue, EV/EBITDA.
SOTP — when 2+ distinct reporting segments exist, value each at pure-play peer multiples.

Always present a WACC × terminal-growth sensitivity table and Bull/Base/Bear scenarios.

Disclaimer: Research/educational output. Not financial advice.

通过三种方法交叉验证内在价值，再综合得出隐含股价：

DCF —— 5年FCFF（自由现金流）预测，以WACC折现，计算终值。
相对估值 —— 应用可比公司中位数P/E、EV/收入、EV/EBITDA倍数。
SOTP —— 当公司存在2个及以上独立报告分部时，使用纯业务可比公司倍数为每个分部估值。

始终呈现WACC × 终值增长率敏感性分析表，以及乐观/基准/悲观场景。

免责声明：本内容为研究/教育用途，不构成财务建议。

Step 1: Detection Flow

步骤1：检测流程

Detect data source and runtime deps. The skill supports 3 method paths — pick the richest one available.

Environment status:

!`python3 -c "import yfinance, numpy, pandas; print('YFIN_OK')" 2>/dev/null || echo "YFIN_MISSING"`

!`(command -v funda && funda --version) 2>/dev/null || echo "FUNDA_CLI_MISSING"`

!`python3 -c "import yfinance as yf; t=yf.Ticker('^TNX'); p=t.fast_info.last_price; print(f'RF_10Y={p/100:.4f}')" 2>/dev/null || echo "RF_FETCH_FAIL"`

Decision tree:

Condition	Method path
`YFIN_OK`	Path A (primary): yfinance for financials + peer multiples
`YFIN_MISSING` but `FUNDA_CLI_MISSING` is not set	Path B: delegate to `finance-data-providers:funda-data` skill for fundamentals
Both missing	Path C: pip-install yfinance, then Path A. `python3 -m pip install -q yfinance numpy pandas`
`RF_FETCH_FAIL`	Use default `rf = 0.045` and note stale risk-free rate in output

RF_10Y=

printed, use that value as

rf

in Step 4d instead of the hardcoded 4.5%.

检测数据源和运行依赖。本技能支持3种方法路径——选择可用的最丰富路径。

环境状态：

!`python3 -c "import yfinance, numpy, pandas; print('YFIN_OK')" 2>/dev/null || echo "YFIN_MISSING"`

!`(command -v funda && funda --version) 2>/dev/null || echo "FUNDA_CLI_MISSING"`

!`python3 -c "import yfinance as yf; t=yf.Ticker('^TNX'); p=t.fast_info.last_price; print(f'RF_10Y={p/100:.4f}')" 2>/dev/null || echo "RF_FETCH_FAIL"`

决策树：

条件	方法路径
`YFIN_OK`	路径A（首选）：使用yfinance获取财务数据和可比公司倍数
`YFIN_MISSING` 但 `FUNDA_CLI_MISSING` 未设置	路径B：委托 `finance-data-providers:funda-data` 技能获取基本面数据
两者均缺失	路径C：pip安装yfinance，然后使用路径A。 `python3 -m pip install -q yfinance numpy pandas`
`RF_FETCH_FAIL`	使用默认值 `rf = 0.045` ，并在输出中注明无风险利率可能过时

如果输出

RF_10Y=

，则在步骤4d中使用该值作为

rf

，而非硬编码的4.5%。

Step 2: Choose Methods & Set Defaults

步骤2：选择方法与设置默认值

Method applicability

方法适用性

Company type	DCF	Relative	SOTP	Fallback
Mature cash-flow (CPG, telecom, utilities)	✅ primary	✅	❌	—
High-growth SaaS / software	✅ with care	✅ primary	❌	Use EV/Revenue + Rule of 40
Multi-segment conglomerate	✅	✅	✅ primary	See `references/sotp.md`
Banks / insurance	❌	✅ (P/B, P/TBV)	❌	DDM or excess return; note in output
Pre-revenue	❌	EV/Revenue only	❌	Flag low confidence
REITs	❌	✅ (P/FFO, P/AFFO)	❌	NAV-based
Cyclicals (energy, semis, industrials)	✅ on mid-cycle	✅	sometimes	Normalize through-cycle

公司类型	DCF	相对估值	SOTP	备选方案
成熟现金流型（消费品、电信、公用事业）	✅ 首选	✅	❌	—
高增长SaaS/软件	✅ 谨慎使用	✅ 首选	❌	使用EV/收入 + 40法则
多分部企业集团	✅	✅	✅ 首选	参见 `references/sotp.md`
银行/保险	❌	✅（P/B、P/TBV）	❌	股息贴现模型（DDM）或超额收益法；在输出中注明
未盈利	❌	仅使用EV/收入	❌	标记低置信度
房地产投资信托（REITs）	❌	✅（P/FFO、P/AFFO）	❌	基于资产净值（NAV）
周期性行业（能源、半导体、工业）	✅ 基于周期中期	✅	有时适用	跨周期标准化

Defaults table

默认参数表

Every parameter below MUST have a value before moving to Step 3. Use these unless the user overrides.

Parameter	Default	Rationale
Projection horizon	5 years	Standard explicit forecast window
Terminal growth `g`	2.5%	~ long-run US GDP
Risk-free rate `rf`	Live 10Y UST from Step 1, else 4.5%	Current cost of capital anchor
Equity risk premium `erp`	5.5%	Damodaran mid-range
Beta	`info['beta']` from yfinance	Market-observed levered beta
Cost of debt `kd`	`interest_expense / total_debt` , else 5.5%	Effective rate; fallback to IG spread
Tax rate	3-yr median effective rate, floored 15%, capped 30%	Strips out one-offs
Margin assumptions	3-yr median of each ratio	Smooths cyclical noise
SBC treatment	Cash for software/SaaS; non-cash for industrials/CPG	Industry convention
Peer count	4-6	Balances signal vs noise
Peer multiple	Median (not mean)	Robust to outliers
Method weights (no SOTP)	DCF 50% / Relative 50%	Equal triangulation
Method weights (with SOTP)	DCF 40% / Relative 30% / SOTP 30%	SOTP gets weight when applicable
Sensitivity grid	WACC ±1% in 0.5% steps × g from 1.5-3.5% in 0.5%	5×5 matrix

See

references/wacc_erp_rates.md

for current risk-free rates, ERP tables, and sector WACC benchmarks.

以下每个参数在进入步骤3前必须有值。除非用户覆盖，否则使用以下默认值。

参数	默认值	依据
预测期限	5年	标准明确预测窗口
终值增长率 `g`	2.5%	~美国长期GDP增长率
无风险利率 `rf`	步骤1获取的实时10年期美国国债收益率，否则为4.5%	当前资本成本锚点
股权风险溢价 `erp`	5.5%	Damodaran中位数区间
Beta值	yfinance的 `info['beta']`	市场观测的杠杆Beta
债务成本 `kd`	`利息支出 / 总债务` ，否则为5.5%	有效利率；备选为投资级利差
税率	3年中位数有效税率，下限15%，上限30%	剔除一次性因素
利润率假设	各比率的3年中位数	平滑周期性波动
股权激励（SBC）处理	软件/SaaS视为现金支出；工业/消费品视为非现金支出	行业惯例
可比公司数量	4-6家	平衡信号与噪音
可比公司倍数	中位数（而非均值）	对异常值更稳健
方法权重（无SOTP）	DCF 50% / 相对估值 50%	均等交叉验证
方法权重（含SOTP）	DCF 40% / 相对估值 30% / SOTP 30%	适用时给予SOTP相应权重
敏感性分析网格	WACC ±1%，步长0.5% × 增长率1.5-3.5%，步长0.5%	5×5矩阵

参见

references/wacc_erp_rates.md

获取当前无风险利率、ERP表格及行业WACC基准。

Step 3: Pull Data

步骤3：拉取数据

python

import yfinance as yf
import numpy as np
import pandas as pd

TICKER = "AAPL"  # replace
t = yf.Ticker(TICKER)

info       = t.info
income_a   = t.income_stmt
cashflow_a = t.cashflow
balance_a  = t.balance_sheet
income_q   = t.quarterly_income_stmt
cashflow_q = t.quarterly_cashflow

earnings_est = t.earnings_estimate
revenue_est  = t.revenue_estimate

price       = info.get("currentPrice") or info.get("regularMarketPrice")
market_cap  = info.get("marketCap")
shares_out  = info.get("sharesOutstanding")
total_debt  = info.get("totalDebt") or 0
cash        = info.get("totalCash") or 0
beta        = info.get("beta") or 1.0
sector      = info.get("sector")
industry    = info.get("industry")

Key financial statement rows (yfinance labels):

Need	Row
Revenue	`Total Revenue`
EBIT	`Operating Income`
Net income	`Net Income`
D&A	`Depreciation And Amortization` (in cashflow)
CapEx	`Capital Expenditure` (negative)
ΔNWC	`Change In Working Capital` (cashflow)
SBC	`Stock Based Compensation` (cashflow)

python

import yfinance as yf
import numpy as np
import pandas as pd

TICKER = "AAPL"  # 替换为目标股票代码
t = yf.Ticker(TICKER)

info       = t.info
income_a   = t.income_stmt
cashflow_a = t.cashflow
balance_a  = t.balance_sheet
income_q   = t.quarterly_income_stmt
cashflow_q = t.quarterly_cashflow

earnings_est = t.earnings_estimate
revenue_est  = t.revenue_estimate

price       = info.get("currentPrice") or info.get("regularMarketPrice")
market_cap  = info.get("marketCap")
shares_out  = info.get("sharesOutstanding")
total_debt  = info.get("totalDebt") or 0
cash        = info.get("totalCash") or 0
beta        = info.get("beta") or 1.0
sector      = info.get("sector")
industry    = info.get("industry")

关键财务报表行（yfinance标签）：

需求	行标签
收入	`Total Revenue`
EBIT（息税前利润）	`Operating Income`
净利润	`Net Income`
折旧与摊销	`Depreciation And Amortization` （现金流表）
资本支出	`Capital Expenditure` （负值）
营运资金变化	`Change In Working Capital` （现金流表）
股权激励	`Stock Based Compensation` （现金流表）

Step 4: DCF Build

步骤4：构建DCF模型

Full methodology + industry-specific tweaks in

references/dcf.md

. Quick skeleton:

python

undefined

完整方法论及行业特定调整参见

references/dcf.md

。快速框架：

python

undefined

4a. Revenue growth path — fade from Y1 (consensus or hist CAGR) to terminal g

4a. 收入增长路径 —— 从第1年（一致预期或历史CAGR）逐步衰减至终值增长率g

hist_cagr = (rev[-1] / rev[0]) ** (1 / (len(rev)-1)) - 1 y1 = float(revenue_est.loc["+1y", "growth"]) if "+1y" in revenue_est.index else hist_cagr g_terminal = 0.025 growth_path = np.linspace(y1, g_terminal + 0.01, 5)

4b. Margins — 3y median

4b. 利润率 —— 3年中位数

ebit_margin = float((income_a.loc["Operating Income"] / income_a.loc["Total Revenue"]).iloc[:3].median()) da_pct = float((cashflow_a.loc["Depreciation And Amortization"] / income_a.loc["Total Revenue"]).iloc[:3].median()) capex_pct = float((cashflow_a.loc["Capital Expenditure"].abs() / income_a.loc["Total Revenue"]).iloc[:3].median()) nwc_pct = float((cashflow_a.loc["Change In Working Capital"].abs() / income_a.loc["Total Revenue"]).iloc[:3].median()) tax_rate = max(0.15, min(0.30, 0.21)) # use effective if available

4c. FCFF per year

4c. 每年FCFF

rev_t = [float(income_a.loc["Total Revenue"].iloc[0])] fcff = [] for g in growth_path: rev_t.append(rev_t[-1] * (1 + g)) ebit = rev_t[-1] * ebit_margin nopat = ebit * (1 - tax_rate) fcff.append(nopat + rev_t[-1]*da_pct - rev_t[-1]*capex_pct - rev_t[-1]*nwc_pct)

4d. WACC

4d. WACC计算

rf, erp, kd = 0.045, 0.055, 0.055 # override rf with live value from Step 1 ke = rf + beta * erp e_v = market_cap / (market_cap + total_debt) d_v = 1 - e_v wacc = e_vke + d_vkd*(1 - tax_rate)

rf, erp, kd = 0.045, 0.055, 0.055 # 使用步骤1的实时值覆盖rf ke = rf + beta * erp e_v = market_cap / (market_cap + total_debt) d_v = 1 - e_v wacc = e_vke + d_vkd*(1 - tax_rate)

4e. Terminal value — compute both, use midpoint

4e. 终值 —— 计算两种方法，取中间值

tv_gordon = fcff[-1] * (1 + g_terminal) / (wacc - g_terminal) tv_exit = (rev_t[-1] * ebit_margin + rev_t[-1] * da_pct) * 15 # peer median EV/EBITDA tv_base = 0.5 * (tv_gordon + tv_exit)

tv_gordon = fcff[-1] * (1 + g_terminal) / (wacc - g_terminal) tv_exit = (rev_t[-1] * ebit_margin + rev_t[-1] * da_pct) * 15 # 可比公司中位数EV/EBITDA tv_base = 0.5 * (tv_gordon + tv_exit)

4f. Bridge to equity

4f. 从企业价值到股权价值的转换

pv_fcff = sum(f / (1+wacc)**(i+1) for i, f in enumerate(fcff)) pv_tv = tv_base / (1+wacc)**5 ev = pv_fcff + pv_tv equity = ev + cash - total_debt implied_price_dcf = equity / shares_out


**Gates:** (a) if `wacc <= g_terminal` → stop, g too aggressive; (b) if `pv_tv / ev > 0.85` or `< 0.45` → flag and show both TV methods; (c) if `wacc` is outside the sector sanity band in `references/wacc_erp_rates.md` → note.

---

pv_fcff = sum(f / (1+wacc)**(i+1) for i, f in enumerate(fcff)) pv_tv = tv_base / (1+wacc)**5 ev = pv_fcff + pv_tv equity = ev + cash - total_debt implied_price_dcf = equity / shares_out


**校验规则**：(a) 如果`wacc <= g_terminal` → 终止，增长率过于激进；(b) 如果`pv_tv / ev > 0.85`或`< 0.45` → 标记并展示两种终值计算方法；(c) 如果`wacc`超出`references/wacc_erp_rates.md`中的行业合理区间 → 注明。

---

Step 5: Relative Valuation

步骤5：相对估值

Select 4-6 peers. Peer map and adjustment rules in

references/relative_valuation.md

python

PEERS = ["MSFT", "ORCL", "CRM", "NOW", "SAP", "WDAY"]  # pick by industry
multiples = {}
for p in PEERS:
    pi = yf.Ticker(p).info
    multiples[p] = {
        "pe_fwd": pi.get("forwardPE"),
        "ev_rev": pi.get("enterpriseToRevenue"),
        "ev_ebitda": pi.get("enterpriseToEbitda"),
        "ps": pi.get("priceToSalesTrailing12Months"),
    }
med_pe     = np.nanmedian([v["pe_fwd"] for v in multiples.values()])
med_ev_rev = np.nanmedian([v["ev_rev"] for v in multiples.values()])
med_ev_eb  = np.nanmedian([v["ev_ebitda"] for v in multiples.values()])

eps_ttm    = float(income_q.loc["Diluted EPS"].iloc[:4].sum())
rev_ttm    = float(income_q.loc["Total Revenue"].iloc[:4].sum())
ebitda_ttm = float(income_q.loc["EBIT"].iloc[:4].sum()) + float(cashflow_q.loc["Depreciation And Amortization"].iloc[:4].sum())
net_debt   = total_debt - cash

implied_pe       = med_pe * eps_ttm
implied_ev_rev   = (med_ev_rev * rev_ttm - net_debt) / shares_out
implied_ev_ebit  = (med_ev_eb  * ebitda_ttm - net_debt) / shares_out
implied_price_rel = np.nanmedian([implied_pe, implied_ev_rev, implied_ev_ebit])

Adjust peer median ±10-30% if target's growth or margin profile diverges materially. Always state the adjustment and reason. Rule of 40 anchor for SaaS in

references/relative_valuation.md

选择4-6家可比公司。可比公司映射及调整规则参见

references/relative_valuation.md

。

python

PEERS = ["MSFT", "ORCL", "CRM", "NOW", "SAP", "WDAY"]  # 按行业选择
multiples = {}
for p in PEERS:
    pi = yf.Ticker(p).info
    multiples[p] = {
        "pe_fwd": pi.get("forwardPE"),
        "ev_rev": pi.get("enterpriseToRevenue"),
        "ev_ebitda": pi.get("enterpriseToEbitda"),
        "ps": pi.get("priceToSalesTrailing12Months"),
    }
med_pe     = np.nanmedian([v["pe_fwd"] for v in multiples.values()])
med_ev_rev = np.nanmedian([v["ev_rev"] for v in multiples.values()])
med_ev_eb  = np.nanmedian([v["ev_ebitda"] for v in multiples.values()])

eps_ttm    = float(income_q.loc["Diluted EPS"].iloc[:4].sum())
rev_ttm    = float(income_q.loc["Total Revenue"].iloc[:4].sum())
ebitda_ttm = float(income_q.loc["EBIT"].iloc[:4].sum()) + float(cashflow_q.loc["Depreciation And Amortization"].iloc[:4].sum())
net_debt   = total_debt - cash

implied_pe       = med_pe * eps_ttm
implied_ev_rev   = (med_ev_rev * rev_ttm - net_debt) / shares_out
implied_ev_ebit  = (med_ev_eb  * ebitda_ttm - net_debt) / shares_out
implied_price_rel = np.nanmedian([implied_pe, implied_ev_rev, implied_ev_ebit])

如果目标公司的增长或利润率与可比公司存在显著差异，调整可比公司中位数±10-30%。始终说明调整及原因。SaaS行业的40法则锚点参见

references/relative_valuation.md

。

Step 6: SOTP (multi-segment only)

步骤6：SOTP估值（仅多分部公司）

Skip unless the 10-K reports 2+ operating segments with distinct economics. yfinance does NOT expose segment data — user must supply or parse from filings. Full methodology in

references/sotp.md

Identify segments + pure-play peer for each
Apply peer median EV/EBITDA (or EV/Rev for growth segments)
Subtract unallocated corporate costs (cap 2-5% of revenue if unknown)
Subtract net debt, minority interest; divide by shares

SOTP discount = (SOTP price − market price) / SOTP price. Flag if >20% (conglomerate discount).

除非10-K报告显示存在2个及以上具有独立经济特性的运营分部，否则跳过此步骤。yfinance不提供分部数据——用户必须提供或从文件中解析。完整方法论参见

references/sotp.md

：

识别各分部及对应的纯业务可比公司
应用可比公司中位数EV/EBITDA（增长分部使用EV/收入）
扣除未分配的公司成本（未知时上限为收入的2-5%）
扣除净债务、少数股东权益；除以总股数

SOTP折价 = (SOTP股价 − 市价) / SOTP股价。如果折价>20%则标记为企业集团折价。

Step 7: Triangulate, Sensitivity, Scenarios

步骤7：交叉验证、敏感性分析与场景模拟

python

undefined

python

undefined

Blended implied price

综合隐含股价

if sotp_price is None: blended = 0.5implied_price_dcf + 0.5implied_price_rel else: blended = 0.4implied_price_dcf + 0.3implied_price_rel + 0.3*sotp_price

5x5 sensitivity grid

5x5敏感性分析网格

wacc_grid = [wacc + dx for dx in (-0.01, -0.005, 0, 0.005, 0.01)] g_grid = [0.015, 0.020, 0.025, 0.030, 0.035] sens = {} for w in wacc_grid: for g in g_grid: tv = fcff[-1]*(1+g)/(w-g) pv = sum(f/(1+w)**(i+1) for i,f in enumerate(fcff)) + tv/(1+w)**5 sens[(w,g)] = (pv + cash - total_debt) / shares_out


Also produce Bull / Base / Bear: shift revenue growth ±300bps, EBIT margin ±200bps, WACC ∓100bps, terminal g 3.0% / 2.5% / 1.5%.

---


同时生成乐观/基准/悲观场景：收入增长率±300个基点，EBIT利润率±200个基点，WACC∓100个基点，终值增长率3.0% / 2.5% / 1.5%。

---

Step 8: Respond to the User

步骤8：响应用户

Output in this order:

Headline verdict — one sentence: blended fair value, vs. current, % upside/downside, most bullish/bearish method. Example: "AAPL fair value ≈ $215 (blended), vs. current $198 → ~9% upside; DCF is most bullish at $228."
Snapshot — sector, industry, market cap, current price, 3M / 12M price change, LTM revenue growth.
Three-method summary — 3-column table: method | implied price | weight | brief rationale.
DCF build — assumptions table (growth path, margins, WACC components, terminal method) + 5-yr FCFF projection table + EV-to-equity bridge.
Peer comparison — table of peers with P/E fwd, EV/Rev, EV/EBITDA, gross margin, rev growth; bottom row = median; flag target's premium/discount.
SOTP (if applicable) — segment table + adjustments + equity value.
Sensitivity matrix — WACC × g grid (5×5), base case highlighted.
Scenarios — Bull / Base / Bear table with levers + implied price.
Key risks — 3-5 bullets: which assumption moves the answer most; what could break the thesis.

按以下顺序输出：

核心结论 —— 一句话：综合公允价值、当前市价、上涨/下跌空间、最乐观/悲观的方法。示例：“AAPL公允价值≈215美元（综合），当前市价198美元→约9%上涨空间；DCF模型最乐观，估值228美元。”
快照信息 —— 行业、板块、市值、当前价格、3个月/12个月价格变动、过去12个月收入增长率。
三种方法汇总 —— 3列表格：方法 | 隐含股价 | 权重 | 简要依据。
DCF模型详情 —— 假设表（增长路径、利润率、WACC构成、终值方法）+ 5年FCFF预测表 + 企业价值到股权价值的转换过程。
可比公司对比 —— 表格展示可比公司的预期P/E、EV/收入、EV/EBITDA、毛利率、收入增长率；最后一行为中位数；标记目标公司的溢价/折价。
SOTP估值（如适用）—— 分部表 + 调整项 + 股权价值。
敏感性矩阵 —— WACC × 增长率网格（5×5），高亮基准场景。
场景模拟 —— 乐观/基准/悲观场景表格，包含调整变量 + 隐含股价。
关键风险 —— 3-5个要点：哪些假设对结果影响最大；哪些因素可能打破估值逻辑。

Error handling

错误处理

Missing / edge case	Action
yfinance returns `None` for beta	Use sector-default beta from `references/wacc_erp_rates.md`
Negative LTM EBITDA	Skip EV/EBITDA multiple; rely on EV/Revenue + DCF
Negative LTM EPS	Skip P/E multiple; use forward P/E if positive, else skip
Growth > WACC in Gordon	Cap `g = wacc − 0.5%` and flag
Fewer than 3 years history	Use what's available; flag data confidence as "low"
Peer data fetch fails	Drop that peer from median; note in output
No segment data for SOTP	Skip Section 6; proceed with DCF + Relative only

缺失/边缘情况	操作
yfinance返回beta为 `None`	使用 `references/wacc_erp_rates.md` 中的行业默认beta值
过去12个月EBITDA为负	跳过EV/EBITDA倍数；依赖EV/收入 + DCF模型
过去12个月EPS为负	跳过P/E倍数；如预期P/E为正则使用，否则跳过
戈登模型中增长率>WACC	将 `g` 上限设为 `wacc − 0.5%` 并标记
历史数据不足3年	使用可用数据；标记数据置信度为“低”
可比公司数据拉取失败	将该公司从中位数计算中剔除；在输出中注明
SOTP估值无分部数据	跳过第6节；仅使用DCF + 相对估值继续

Caveats to include

需包含的提示

TTM data lags real-time; peer multiples reflect market sentiment (can overshoot)
DCF is garbage-in/garbage-out; sensitivity matters more than a point estimate
yfinance data is unofficial; cross-check any decision with primary filings
Not financial advice

TTM数据滞后于实时数据；可比公司倍数反映市场情绪（可能偏离合理区间）
DCF模型结果依赖输入质量；敏感性分析比单点估值更重要
yfinance数据非官方；任何决策需与原始文件交叉验证
不构成财务建议

Reference Files

参考文件

```
references/dcf.md
```
— DCF methodology + industry-specific guidance (software, retail, financials, healthcare, energy, manufacturing, CPG, telecom, REITs, streaming)
```
references/relative_valuation.md
```
— Peer selection, multiple adjustment rules, Rule of 40, peer sets by theme
```
references/sotp.md
```
— Sum-of-parts methodology, conglomerate discount detection, catalysts
```
references/wacc_erp_rates.md
```
— Risk-free rates, equity risk premiums, sector WACC benchmarks, sector-default betas

```
references/dcf.md
```
—— DCF方法论 + 行业特定指南（软件、零售、金融、医疗、能源、制造、消费品、电信、REITs、流媒体）
```
references/relative_valuation.md
```
—— 可比公司选择、倍数调整规则、40法则、按主题划分的可比公司组
```
references/sotp.md
```
—— 分部加总法方法论、企业集团折价检测、催化因素
```
references/wacc_erp_rates.md
```
—— 无风险利率、股权风险溢价、行业WACC基准、行业默认beta值