投资人在股东信、访谈、雪球上公开讲过的观点，加上知名投资节目和名人对个股的看法，按发表时间排列。每条都附原文出处，原话逐字照录；投资人谈自己 13F 持仓的，「说完之后」那一行是同一季度 13F 上的实际操作。
达摩达兰分析 AI 资本开支如何流向芯片、网络设备、电力和数据中心公司，英伟达是最大受益者。他说自己大部分组合选择跟随市场共识，持有七巨头中除特斯拉和英伟达以外的五家，其中亚马逊、Alphabet、Meta、微软都是 AI 资本开支大户；但过去一两年新增资金大多放在收益 4% 的短期国债里，组合现金比平时多，他承认少赚了，但睡得更好。
I am personally going with the "market consensus' choice for the bulk of my portfolio, since I do hold five of the Mag Seven (all except Tesla and Nvidia) and four of my holdings in this group (Amazon, Alphabet, Meta and Microsoft) are heavy investors in AI cap exSpaceX 上市时，达摩达兰估值约每股 100 美元；股价已回落到接近这个价位，他可能很快面对一只被低估的股票。由于估值的不确定性很大，他不会有足够的把握让 SpaceX 成为最大或唯一的持仓，但准备把它作为组合中众多押注之一买入。他借此说明：对把握不大的年轻公司，可以靠分散下注来克服不敢出手的犹豫。
Given the uncertainty that is associated with my estimate, and you can see it in the simulation that I reported in my post, it is unlikely that I would ever have sufficient conviction to make SpaceX the biggest or only investment in my portfolio, but I stand ready to buy the stock as part of a portfolio, where it is one of many bets that I make on markets.SpaceX 招股书公布后，达摩达兰更新了估值：上调发射业务利润率、把 AI 业务目标收入翻倍，但调低了 AI 利润率并放慢近期增长，股权价值约 1.3 万亿美元、每股约 100 美元。他最担心 SpaceX 在 AI 上过度扩张，而马斯克的超级投票权让股东难以制约。按传闻的 1.8 万亿美元定价，他觉得太贵；但他也说，看市场情绪和动量的交易者参与发行可能会赚钱。
At the rumored pricing of $1.8 trillion for the company, it is too richly priced for my tastes, given my valuation of $1.25-$1.35 trillion for the equity in the company.库克卸任之际，达摩达兰回顾说，苹果在库克手下从靠颠覆驱动的增长机器，变成了成熟、持续回馈现金、更谨慎的公司。他在乔布斯和库克时期大部分时间都持有苹果；库克顶住了伊坎加杠杆、艾因霍恩发优先股等激进投资者的压力，也没有追逐增长和自我驱动的收购。他说即使苹果在 AI 上的谨慎最终被证明是错的，他仍感谢这种克制。
While there are many who compare to Tim Cook to Steve Jobs and find him wanting on vision and flair, I am grateful, as an investor in Apple, for the restraint and discipline that he brought to the job.在招股书发布前，达摩达兰用零散信息给 SpaceX 做了首次估值，约 1.2 万亿美元。他认为 SpaceX 是工程奇迹、拥有巨大竞争优势，但像特斯拉一样会在不同叙事间切换，让投资者坐过山车，所以既难买入，做空更危险。他不打算以传闻的 1.75 万亿美元发行估值买入，但认为只要大跌一次就会合理甚至便宜，到时他会买。
With all that said, SpaceX is a unique company with immense competitive advantages, and while I would not be interested in buying at the rumored IPO pricing of $1.75 trillion, it is one big correction away from being fairly priced or even cheap.达摩达兰指出，2025 年金价涨 65%，却既没有高通胀，也没有严重危机；按 CPI 调整后的金价表现远超「黄金长期跟随通胀」的规律。央行增持黄金有影响，但解释不了全部涨幅。他认为股票和债券市场基本没理会信任问题，汇市有些担忧，而金银显示一部分投资者看到了灾难性风险、正在涌入贵金属。
When investors lose trust in governments and central banks, it should come as not surprise that their money leaves financial asset markets and goes into collectibles2025 年美国关税、评级下调、史上最长停摆和美联储独立性受质疑，削弱了市场对美国政府的信任，黄金涨 65%、白银涨 148%。达摩达兰指出比特币全年反而跌了 6.4%（以欧元计跌 17.4%），而且和美股的相关性远高于和金银的相关性，吸引的是追求风险的人，而不是规避风险的人。他认为现在下定论还太早，但至少 2025 年，失去信任的资金去了金银，而不是比特币。
It remains too early in bitcoin's life to pass final judgment, but if the story for bitcoin is that it will draw in investors who have lost trust in governments and central banks, it is clear that gold and silver were the draws, at least in 2025, not bitcoin.2026 年初，达摩达兰算出美股隐含股权风险溢价为 4.23%，几乎等于 1960–2025 年的平均。他认为市盈率偏高值得担心，但盈利增长强劲、现金回报丰厚（2025 年标普 500 公司回购超过一万亿美元），多空双方都有理由。他对「泡沫」一词保持警惕：1999 年末风险溢价低到 2.05%，而现在处在历史区间的中部。
I am leery of the bubble word, especially used in the context of this market, since unlike the end of 1999, when the ERP got as low as 2.05%, the current ERP is more in the middle of the historic range.英伟达市值一度超过 5 万亿美元后，达摩达兰用「保本收入」倒推：要撑起当前市值，一家公司五年后需要做到多少收入。他提醒「定价高」不等于「高估」，但就他自己的判断，支撑英伟达和 OpenAI 当前定价所需的收入增长和利润率不太可能实现；他也说别人得出相反结论完全可能。
I think that the pathway that you would need (in terms of revenue growth and profitability) to justify Nvidia's and OpenAI's current pricing is improbable, but that is just my view, and it is entirely possible that you have an alternate perspective, leading to the conclusion that they are undervalued.2025 年 10 月金价突破每盎司 4000 美元。达摩达兰说自己从没被黄金吸引过，也从没持有过。他把黄金归为收藏品：平时买黄金的多是不信任法币和金融市场的「信徒」，但在动荡时期，原本买股票债券的投资者也会因为不信任央行或担心泡沫而涌入。他认为即使不买黄金的投资者，也不该忽视金价上涨传递的信号。
I must confess that I have never felt the draw of gold, and have never held it in my portfolio, but I have always been fascinated by the hold that gold has on some investors, and the reasons for its longevity.鲍威尔形容股市「估值相当高」，达摩达兰同意美股定价偏贵。但他用席勒 1871–2025 年的数据回测了按 CAPE 择时的策略，认为从「贵」跳到「泡沫」再据此行动要难得多：择时指标越不完整，越可能在等一场不会来的回调；即使做足功课，择时依然充满噪音，市场错误定价的时间可能比你撑得更久。
As someone who buys into the notion that market timing is the impossible dream, you may find it surprising that I think that Powell is right in his assessment that stocks are richly priced, but that said, I will try to explain why making the leap into concluding that stocks are in a bubble, and acting on that conclusion are much more difficult to do.达摩达兰一直认为比特币只能定价、不能估值，买不买取决于你对价格走势的判断。谈到企业把现金换成比特币，他认为对大多数公司来说这不但是坏主意，还应该被禁止：这会搅乱主业的故事线，让财报难以解读，也给管理层自肥打开方便之门。他说即使自己看多比特币也会这么主张，换成黄金、名画也一样。
As a general rule, I think it is not only a bad idea for most companies to invest their cash in bitcoin, but I would go further and also argue that they should banned from doing so.在市场暴跌期间，达摩达兰对三家一直想持有的公司挂了限价买单，其中之一是 Palantir，他认为这是最接近兑现 AI 产品与服务承诺的公司。截至发文，比亚迪已经成交，Palantir 和美客多离他的限价还有距离，他说买单一直有效，直到撤单。
Palantir, a company that I believe is closest to delivering on thee promise of AI products and services达摩达兰把自己归为「机会型逆向投资者」：趁市场暴跌，买入一直想持有、但之前价格太高的公司。他为此对比亚迪、美客多和 Palantir 挂了限价买单，比亚迪在 4 月 7 日短暂跌破 80 美元的限价时成交。他说买入时要顶住同伴压力，自己有长期视角和好胃口，但并非不受影响。
The limit buy kicked in on BYD on April 7, when it briefly dipped below $80, my limit price关税引发市场暴跌后，达摩达兰把自己认为是伟大公司、但此前价格太高的股票重新估值，计入关税影响，并在低于估值的价位挂了限价买单。他形容美客多是拉美巨头，兼具亚马逊式的颠覆潜力和金融科技业务；暴跌一周内，美客多的股价已从明显高估跌到接近他的限价。
Mercado Libre, a Latin American powerhouse, that has the disruptive potential of an Amazon combined with a fintech enterprise特斯拉在大选后大涨，随后几乎回吐全部涨幅，达摩达兰认为主要原因是马斯克的政治关系。他仍认为特斯拉会是电动车的领先者之一，但称霸之路更坎坷：电动车市场可能分化为比亚迪领先的低价市场和特斯拉守住的高端市场。他把估值下调到每股约 150 美元，比一年前低约 30 美元，比股价低约 70 美元；他还说，既然特斯拉已成政治标的，或许是时候考虑比亚迪了。
My estimate of value for Tesla stands at about $150 a share, about $30 less than my value last year, and about $70 below its stock price.年初达摩达兰给标普 500 估值 5262 点，比年初的 5882 点低约 12%。3 月上半月股权风险溢价从 4.35% 升到 4.68%，反映关税和经济的不确定性。他认为风险溢价的走势会决定美股是企稳、继续下跌还是反弹；即使经过 3 月的回调，指数仍高于他的估值，而过去几周发生的事只会拉低而不是抬高合理价值。
Even with the drawdown in prices that we have seen through March 10, the index remains above my estimated value, and while that value reflected what I saw at the start of the year, what has happened in the last few weeks has lowered the fair value, not raised it.DeepSeek 用低成本做出大模型后，达摩达兰认为 AI 市场会分成低投入、易同质化和高投入、高利润两条路，AI 芯片市场因此会比原来预想的小。他把英伟达每股价值从 2024 年 9 月的 87 美元下调到 78 美元，认为 123 美元的股价即使在下跌后仍然高估。他 2023、2024 年夏天已先后卖掉一半和四分之一的英伟达，仍保留一部分。
The other is that the stock is overvalued, at its current price of $123 per share, even after the markdown this week.2025 年初，标普 500 按过去十二个月盈利计算的市盈率为 24.16，高于 1970 年以来每个十年的平均。达摩达兰认为通胀、关税和战争会带来很大波动，他更愿意看到更高的股权风险溢价；但当前溢价不像 1999 年末那样只有 2%，所以他不站在泡沫派一边。他对盈利、利率和风险溢价做了模拟，结论是在当前价位，美股高估的概率是 80%，低估的概率只有 20%。
Put simply, at today's price levels, there is an 80% chance that stocks are overvalued and only a 20% chance that they are undervalued.司法部表示考虑拆分 Alphabet。达摩达兰认为这既困难又不明智：谷歌搜索、YouTube、安卓的核心都是广告，价值在于把用户留在谷歌生态里，拆开后各平台既难以独立生存，对用户的吸引力也会下降；在线广告有网络效应，拆分也许带来更多竞争，却会损害效率，消费者会失去免费的产品，广告投放也更不精准。他认为更可能的结果是要求苹果、谷歌开放平台。
In short, the platforms, separated, would both be unable to survive as stand alone businesses as well as less attractive destinations for users.在讨论品牌老化时，达摩达兰回顾了 2015 年巴菲特联手 3G 资本收购卡夫亨氏：当时他就怀疑，市场看淡这家公司，是因为喜欢它产品的人群越来越少、越来越老。他认为品牌和人一样会老，管理层再怎么努力也很难让老品牌找回活力；事后看，卡夫亨氏的产品和客户群都已老化，不会重回巅峰。
In hindsight, it is clear that Kraft Heinz will not reclaim its former glory, because its products and customer base have aged.达摩达兰认为饮料、麦片这类公司除了品牌几乎没有其他差异，最容易估算品牌价值。他按可口可乐当前 29.73% 的税前营业利润率估值，公司价值约 2811.5 亿美元，略微高估；再把它和软饮料公司 16.75% 的加权平均利润率、全球饮料公司 6.92% 的中位数比较，拆出品牌本身的价值。
Note that while the company comes in as slightly overvalued, it is still given a value of $281.15 billion, with much of that value coming from its pre-tax operating margin of 29.73%.达摩达兰读完《鞋狗》后分析耐克的品牌价值：除 2023 年外，耐克的营业利润率一直比行业高 1.5–3 个百分点，说明品牌有溢价；但转向线上直营、绕开鞋店可能削弱了品牌，2023 年是警钟。他按三个时期的增长和利润率估值，如果增长和利润率回不来，价值会大幅下降。六周前 71 美元时胜算在投资者一边，换帅后反弹到 88 美元，市场已在定价复苏。
At the $71 stock price that the stock was trading at, just six weeks ago, the investing odds would have been in your favor, but the bounce back in the stock price to $88, after the new CEO hire, suggests that the market is pricing in the expectation that the company will bounce back to higher growth and better margins.达摩达兰认为星巴克的收入增长仍然不错、也还赚钱，最大的问题是丢了故事线，需要一位既有愿景、又懂咖啡店运营的掌舵人，他不确定新 CEO Brian Niccol 是否合适。他说用任何历史增长和利润率组合都算不到当前股价，除非星巴克既保持 2002–2011 年的增速，又拿到 2012–2021 年更高的利润率，他认为这不现实。
I must be missing some of the Starbucks magic that investors are seeing, since there is no combination of historical growth/margins that gets me close to the current stock price.达摩达兰认为英特尔的问题在于跟风和追求够不着的增长：AI 芯片对上英伟达、代工对上台积电，都没有领先优势，但市场足够大，它能有一席之地，只是做不了老大。按 2024 年 9 月 8 日 18.89 美元的股价，只要利润率回到正常水平就够了：3% 增长、25% 营业利润率下每股价值已有 23.70 美元。AI 芯片和《芯片法案》的收益是额外的，他正在买入英特尔。
I do believe that Intel will derive some payoff from both, and I am buying Intel, to twin with what is left of my Nvidia investment from six years ago.达摩达兰认为沃尔格林的选择越来越少：收缩到最赚钱的门店也许是出路，但债务必须同步降下来，宏观放缓或资本市场冲击可能让股东血本无归。他按不同增长和利润率假设估值，结论完全取决于营业利润率能否回到 4% 以上的历史水平；而他认为药房业务正面临颠覆，利润率回不到 2021 年前，所以不会买。
I believe that the pharmacy business is ripe for disruption, and that the margins will not revert back to pre-2021 levels, making Walgreens a "no go" for me.英伟达财报收入和盈利都超预期，股价却跌了 8%、蒸发 2000 多亿美元市值。达摩达兰说自己不擅长交易，不判断动量变化，而是借财报更新了对英伟达的估值：把它拆成 AI 芯片、汽车芯片和其余业务（加密、游戏）三块。他认为交易者希望它维持现在的增速，这既不现实也危险；他对 2023 年夏天卖掉一半、保留一半的决定很坦然。
While traders may be disappointed with Nvidia’s earnings release, and wish it could keep its current pace going, I think it is both unrealistic and dangerous to expect it to do so.达摩达兰重新估值了 2023 年撑起大盘的「七巨头」。他说这七家公司他都持有，除特斯拉是上周刚买入的，其余都是多年前觉得低估时买的。按他的估值，特斯拉和 Meta 接近合理价，而微软、英伟达偏贵；他没有卖出高估的部分，原因之一是估值本身有误差，另一个是要交近 30% 的资本利得税。
I own all seven of these companies, which may strike you as contradictory, but with the exception of Tesla that I bought just last week, my acquisitions of the other seven companies occurred well in the past, and reflected my judgments that they were undervalued (at the time).在对七巨头的估值里，达摩达兰认为特斯拉、Meta 接近合理价，Alphabet、苹果、亚马逊离价值不远，微软和英伟达则高估，英伟达尤其明显。他说这两家恰好是最直接受益于 AI 热潮的公司，他的高估结论也许只说明自己想象不出 AI 能做多大，但他已经在英伟达的估值里计入了大量 AI 价值。
It may be coincidence, but these are the two companies that have benefited most directly from the AI buzz, and my findings of over valuation may just reflect my lack of imagination on how big AI can get as a business.在七巨头估值更新里，达摩达兰假设 Meta 未来五年收入年均增长 12%、目标营业利润率 40%，算出每股价值 445.10 美元，股价 456.08 美元，只高估约 2.5%，和特斯拉一样接近合理价。他持有七巨头全部七家，没有卖出略贵的几家，原因之一是估值本身有误差，价格还在合理区间内，另一个是要交近 30% 的资本利得税。
Tesla and Meta look close to fairly valued, at current prices达摩达兰假设 Alphabet 未来五年收入年均增长 8%、目标营业利润率 30%，每股价值 138.14 美元，股价 145 美元，高估约 5%；苹果（价值 176.79 美元、股价 188 美元）、亚马逊（价值 155.72 美元、股价 169.15 美元）也都在估值附近。他提醒，公司好不好和股票便宜不便宜是两回事，投资是找你对公司的判断和市场定价之间的错配。
Alphabet, Apple and Amazon are within striking distance of value达摩达兰用 2023 年的经营数据更新了七巨头估值。他假设微软未来五年收入年均增长 15%、目标营业利润率 45%，算出每股价值 355.88 美元，而股价是 405.49 美元，高估约 14%。他认为微软和英伟达是七巨头里最高估的两家，恰好也是最直接受益于 AI 热潮的公司；他已经给了微软因 AI 带来的更高增长。他多年前在低估时买入微软，目前仍持有。
Microsoft and Nvidia look over valued, with the latter especially so.2024 年初，达摩达兰用分析师一致预期的盈利和当时的利率给标普 500 估值，结论是美股高估约 9.2%，而 2023 年初还略显低估。他提醒这个估值有两个风险：经济未必软着陆，盈利可能低于或高于预期；市场寄望 2024 年降息，但十年期国债利率更多取决于通胀而不是美联储。他说自己不擅长择时，不会因此买指数看跌期权或降低股票仓位。
Unlike early in 2023, when stocks looked slightly under valued, with consensus earnings numbers and prevailing rates, stocks look over valued by about 9.2%, with a similar structure today.达摩达兰把特斯拉拆成汽车、软件、机器人出租车等几块业务分别估值，合计每股约 180 美元，高于年初的 153 美元，主要因为副业的利润潜力更高。按当时 197 美元的股价，他认为仍然高估，但股价几周内已跌了 54 美元，可能再跌 20 美元，所以在 180 美元挂了限价买单。
At $197 a share, Tesla remains over valued, at least based on my story, but a stock that has dropped $54 in price in the last few weeks could very well drop another $20 in the next few. To capture that possibility, I have a limit buy at my estimated value of $180, with the acceptance that it may never hit that price in this iteration.Birkenstock 上市前，达摩达兰把品牌、管理层等无形资产都算进增长、利润率和风险的假设里，估出股权价值约 83.8 亿欧元，比传闻的发行定价低 10% 左右。他认为这是一家品牌独一无二、管理层懂得用好品牌又不过度消耗的公司，但 45–50 欧元的发行价偏高；如果市场转差、股价跌到 40 欧元以下，他会买入。
It is, in my view, a reach at €45 or €50 per share, but if the market turns sour, and the stock drops to below €40, I would be a buyer.