Introduction

Anchoring bias—the tendency to rely too heavily on initial information—affects analyst price targets. When stocks rise sharply, analysts adjust targets slowly, remaining anchored to prior levels. Machine learning NLP can detect anchoring by analyzing target-setting language and comparing targets to fundamental valuations, identifying mispricings.

NLP Language Analysis

Parse analyst reports and extract language around price target setting. Phrases like "we maintain our target" (anchored) versus "we significantly raise our target" (adjusting) signal anchoring. Measure target adjustment lag relative to fundamental changes. Machine learning quantifies anchoring magnitude across analysts and stocks.

Misprice Identification

Stocks with high anchoring (targets significantly below current prices and fundamentals) are undervalued; stocks with targets elevated due to past momentum are overvalued. Exploit these mispricings by going long undervalued anchored stocks.

Conclusion

NLP-detected anchoring bias reveals systematic analyst forecast errors, enabling alpha generation.