Home › Market › 2026-10-19

US Stock Analysis and Analyst Forecasts — October 19, 2026

Tech bounced back as the Nasdaq 100 rose about 1.2% and the S&P 500 about 0.4%; Amazon jumped about 5% on a Morgan Stanley target hike to $350, Nvidia gained 2%, and J.P. Morgan upgraded Intel to Neutral while the 10-year yield hit 5.14%.

None of the following is our own view on any stock — it is a same-day record of market action, analyst-firm calls and company news across the tracked universe on Monday, October 19, 2026.

U.S. stocks closed higher as technology bounced back from last week's selloff and oil eased. The S&P 500 gained about 0.4% to 7,691.39, the Nasdaq 100 rose about 1.2% to 29,529.34 and the Dow Jones Industrial Average slipped about 0.07% to 51,966.03. Consumer discretionary led the sectors with a gain of about 2.7%, while utilities lagged with a loss of nearly 2% as the 10-year Treasury yield rose about four basis points to 5.14%, a multi-decade high. Brent crude edged down about 0.5% to roughly $109 a barrel after a report of progress in U.S.-Iran talks. The session fell on the 39th anniversary of the 1987 Black Monday crash.

Amazon (AMZN) was the day's standout, rising about 5% after Morgan Stanley analyst Brian Nowak reiterated a Buy rating, raised his price target to $350 from $335 and named the stock a top pick. He pointed to faster growth at AWS, helped by a reported agreement to buy about 2 million Nvidia GPUs, along with AI shopping assistants and warehouse robotics. He models AWS revenue growth of 47% in 2027 and 42% in 2028. Amazon's consensus is Strong Buy from 42 Buy and two Hold ratings, with an average target of $330.49.

Chip names recovered after a rough week in which hedge funds, according to Goldman Sachs data, cut technology positions at the fastest pace since at least 2016 in the five sessions through October 15, with semiconductors accounting for the largest dollar amount of selling. Micron (MU), Intel (INTC) and Palantir (PLTR) closed with notable gains. Nvidia (NVDA) rose about 2% to roughly $218 after announcing what it called a record fault-tolerant quantum computing simulation, run on its GH200 chips at Germany's Jülich Supercomputing Center; its consensus is Strong Buy from 33 Buy ratings with an average target of $324.20.

Two analyst calls stood out in semiconductors. Deutsche Bank's Melissa Weathers raised her Lam Research (LRCX) price target to $375 from $320 and kept a Buy rating ahead of the company's earnings on Wednesday, October 21, when analysts expect revenue up about 53% from a year earlier to roughly $8.14 billion. Separately, J.P. Morgan's Harlan Sur upgraded Intel to Neutral from Underweight with an $85 target, which sits below the current price, noting that Intel's server CPU share has fallen to 53% even as its manufacturing has improved; he kept Marvell (outside our tracked universe) at Overweight with a $360 target. Intel's consensus is Hold.

In retail, Truist's Scot Ciccarelli reaffirmed a Hold rating on Costco (COST) with a $955 price target, about 3% below Friday's close of $989.11, saying the company keeps posting the best same-store sales growth among major retailers but trades at roughly 42 times earnings. Costco's September net sales rose 13% to $30.02 billion. Apple (AAPL) said it will hold a product event on October 28, while Jefferies warned that demand for the iPhone 18 Pro and Pro Max is weakening, citing shorter delivery times and falling resale prices.

The earnings calendar is busy this week. Netflix (NFLX) reports Tuesday after the close, along with United Airlines and Capital One, which are outside our tracked universe. On Wednesday, Tesla (TSLA), Meta Platforms (META), IBM (IBM), Texas Instruments (TXN) and Lam Research are due to report.

As always: a price target is one firm's estimate rather than a forecast of what a stock will do, a single upgrade or downgrade can differ sharply from the consensus, and none of this is a recommendation to buy or sell anything.

Tickers in this brief

Published 2026-10-19. Market commentary, not investment advice. Prices are delayed end-of-day data.