lnofeisone

Starters in INTC and COF.

Can you share more please, especially around the tax structure (I always love clunky tax structures)?

There is some fun history here - and this is my electrical engineer side. X Fab looks to produce analog devices. Those generally tend to run larger than digital chips. The sizing they give you corresponds to the actual sizing. 3nm stuff is marketing fluff, and those chips are much bigger than 3nm. So it's not apples to apples, but yes, the analog world is much larger. In the analog world, 1um is not terribly far behind "cutting edge." Take a look at at ADI's roadmap and zoom in on the sizing. https://www.analog.com/media/en/news-marketing-collateral/solutions-bulletins-brochures/adis-resilient-hybrid-manufacturing-network.pdf The bit on history: When semiconductors were first produced the sizing was describing the size of the transistor gate length. Over time because of the chip architecture innovation, "nm" took on a marketing tilt. The 3nm doesn't mean a 3nm transistor gate length. All it means is that 3nm process makes smaller stuff than a 7nm process of the same company. The latter point is important because it means that 7nm process of one company can be bigger or smaller than 7nm process of another company.

I think there is a lot of truth to what Einhorn is saying and we see it play out with multiple favorites on this board - JOE, BTI, FRPH, CPNG. One of my holdings - VET. Small companies where value is "easier" to recognize are appreciating just fine without shareholder yield. Companies where things are complex really do need that extra push. The value is not dead, it just requires a little bit more thinking, recognition, and patience.

This. Demand for NatGas is notoriously hard to model. It's subject to weather and, for the longest time, it was centered around local markets. This is why LNG imports were hyper-profitable. Now that LNG transportation infra is becoming more robust, prices across the world will harmonize better, i.e., there won't be a $20 difference between NA and Asian NatGas but there will still be a difference to account for shipping cost/time. Supply is a bit easier to model as you can assume companies will pump out as long as their profit is above or at $0.

Depends on the industry you are in. I've supported model development for several CFO shops. We usually had ranges, and then scenarios planned, i.e., if this happens (e.g., your workforce quits or demands higher pay), then the range gets adjusted like so and so. The complexity of models goes up with more business lines, geographic diversity, etc. At some point, models become more directional or you have a forest of sub-models that average out to be generally correct. Really hard to model discontinuities like rapid interest rate hikes, COVID, Ukraine/Russia war, etc.

You have to deconstruct the problem. Usually LLMs are trained with question-answer pairings (take a look at SQuAD). It's what data you use and how you decompose it into Q-A pairs. Court cases tend to have a lot of reasons AND (this the core value of court data) you have a quantitative measurement of truth. For example, Party A sues Party B and asks for $1,000. If Party A gets $1000, you can label their argument as correct and Party B as incorrect (I'm grossly oversimplifying here). Party A gets $500, you can label it as half correct, etc. You can refine your vector database with assigned probabilities and use that to do dynamic Q-A build-outs. This works really well for general public use cases such as landlord disputes, small claims, etc. Doesn't work as well for big one-of-a-kind cases that will have major discontinuity (e.g., Roe v. Wade, Plessy v Ferguson -- Brown vs. Board of Ed). I would also hire a real lawyer if I have serious $$$ on the line. ChatGPT is trained with a lot of data, some of which can be questioned all day long. Court data is FAR better. It gets even better once you take into account metadata (e.g., timelines, law firms, etc.). Legal language is unique, though decisions have a bit of narrative, and can be somewhat arbitrary. General language models like ChatGPT or Gemini will not do well (due to hallucination for example) but if you decompose it into sub-models your results vastly improve. You can even get substantial result improvements with basic fine-tuning. If you can, play around with BloombergGPT. Few Gov't agencies are also building out their own LLMs, and the results are substantially better than ChatGPT/BARD, which is expected as those models are specialized (i.e., you'll get HORRIBLE answers if you ask something general but amazing results that are specific to the domain of the model).

Judges and clerks who wrote decisions are your quality control. Lots of work has already been done by them, and incorporating them into your learning loop is essentially free.

Great idea but it exists already - it's called Retrieval-Augmented Generation. There is also a heavy shift from general closed models (think chatGPT) to more open (so web + research articles) to more domain specific (e.g., specific area of law, specific area of tax code, etc.). You also don't need a team of paralegals and lawyers if you are trying to help the "average Joe." Just need court data (verdicts, etc.) which is generally easy to get. Where you need a team of lawyers and actual law firms is some of those private settlements. I know of one law firm that has experimented with that but with mixed results. I have theories on why they had mixed results and happy to share them for a substantial fee (in case someone on this board is from that firm). Broadly speaking, however, AI is driving a lot of investments at the corporate level. Many of my clients are finding money for pilots, prototypes, and near-production-ready solution testing. Where a lot of them hit the pause button is when they realize that they need to revamp the entirety of their data ecosystems and project estimates go 10-100x but the willingness to experiment hasn't abated.

I think there is a squeeze in the shipping pricing but not in the cost of energy. US has enough natty. Canada has enough natty. Europe is oversupplied for this time in the winter. I think the prices will continue to stay where they are short to mid term.

I do this a lot with credit cards 0% BT (there is a 3-4% fee). Mostly because I enjoy it. One suggestion I can make is buying a CD through brokerage reduces the extra paperwork you have to do. Treasuries are fine through TreasuryDirect or brokerage.

NNI, NEP. Built full position in both over the last two weeks. The fundamentals on both are either underappreciated (NNI) or assigned too much risk (NEP).

Here we are clocking in with 1400 sq ft for a family of soon to be 4 + dog. Smallish backyard, smallish front yard. Things are a bit tight. We went with a balance of location/affordability/reduction in maintenance. So far it's worked well. We do have 700 sq ft in the basement that are being remade into an apartment. We can "grow into" the apartment at some point but for right now the plan is to rent it out. We might also just go ahead and get another house. The rent for upstairs + downstairs can get us 2x our mortgage payment.

Nobody is being violent. It is an expression. As far as your option trades, I think like anything you need tonfind your edge and execute it. Sounds like you learned a lot but your edge is in finding and holding compounders. That doesn't make options useless.

I'm in violent disagreement with your take—anything from hedging to the downside to minimizing capital needed to place a bet (i.e., lever up). Tax can be well managed with LEAPs. I would agree that being an options trader is definitely a hard day job, but the instrument is immensely valuable.

Thanks for this. 1) He validates half my research that if you buy at a discount at issue (that's the nuance), that extra income is tax-free. 2) Where he invalidated my hypothesis is the secondary market bond purchase. Buying bonds in the secondary triggers issues that would be consistent with other bonds. I found relevant IRS passage that speaks to this (https://www.irs.gov/publications/p550#en_US_2022_publink100010016 - go to Market Discount Bonds). You still get quirk on 30 years out muni zeros. For example: 30 years out, 1000 par, issued at $250 can be bought for $175 ($250-30*0.0025*1000) before triggering any tax issues. In other words, $75 allowance has more mileage on a $250 (30%) price than it would on $950 (8%) bond.

It took a hot minute, but I made a few calls to IRS-savvy individuals. 1) They were equally perplexed at not running into this situation. Our current hypothesis is that none of us have encountered this because we have never had a material interest rate increase + investment in munis at the same time. 2) Onto more interesting tidbit. Original Issue Discount (that's the language IRS calls this) is not included in income if the bond is a tax-exempt obligation. That means that de minmus rule generally does not apply to tax-exempt obligations. Page 6 of the https://www.irs.gov/pub/irs-pdf/p1212.pdf section on "including OID in income" and "De minimis rule" sections provide two very explicit rules on excluding this income. So the opportunity is - buy munis that re below de minimus and, if you hold to maturity, all the gains are tax free. @Dinar - I appreciate your skepticism and would love if you took a look. I bought a bunch of zero munis that are below this de minimus and will wait to see if TD Ameritrade will send the 1099-OID this year.

If you are bullish, which this trade suggests you are, why not go with Jan '26 $50 put? You get more cash upfront and have a greater return if EBAY goes up to $50, and if EBAY drops sub $40, you are only losing marginally more. You can also look at Jan '25 $50 put. You get a bit less cash (still more than your original trade) than the above but you are cutting your put exposure time by 50%. You'll have more flexibility with this positioning come Jan '25, i.e., convert your call to vertical or sell another put.

One of my favorite trades is buy warrants/sell leaps. This trade works particularly well when the stock price exceeds the warrant strike price and the call price starts to be more expensive than leaps for the same duration.

I'm up roughly 20% across the board mostly due to some concentration in the portfolio. My biggest detractor was and continues to be VET. VRRM and SAVE are the biggest gainers. SAVE was a late bloomer as I sold a bunch of puts when it dipped sub $10.

Same. Bought a starter.

Thanks for the counterpoints @Sweet. Forced me to do a bit more digging. This reminds me of VRRM, a very similar setup with tech entrenchment, except here you have diagnostic equipment, which has so many Gov't hurdles. ILMN has a lot of contracts now, and all they have to do is keep up. Easier said than done, I know, but clients aren't keen on switching technology unless the payoff is exponential. The costs get astronomical very fast. Also, this dropped just last month. https://www.nature.com/articles/s41588-023-01540-6 ILMN can also do long-range sequencing - https://www.illumina.com/science/technology/next-generation-sequencing/long-read-sequencing.html. It's a long way of saying that some positives here to make risk/reward with the right sizing.

Aren't the two markets different for ILMN vs. Nanopore. ILMN trumps in accuracy, which makes it better for clinical studies, while Nanopore is great for field studies with its length of sequencing but less accuracy.

Forgot to add mrtx too. I think a lot of biotechnology m&a under 10b will have no issues closing. I have no insights into CVRs but I like having these options.

I really like Illumina. I've also been playing some biotech liquidations, e.g., THRX. I also saw Reneo (RPHM) dropped because they failed but have a decently clean balance sheet with liquidation imminent.